Fable 5 vs GPT-5.6

I think it's finally time. You guys have I think it's finally time. You guys have been asking for my thoughts on Fable been asking for my thoughts on Fable been asking for my thoughts on Fable versus Soul for a while. And now that versus Soul for a while. And now that versus Soul for a while. And now that it's been long enough, I think I finally it's been long enough, I think I finally it's been long enough, I think I finally am ready to talk about it. I've had am ready to talk about it. I've had am ready to talk about it. I've had access to Soul for quite a bit since access to Soul for quite a bit since access to Soul for quite a bit since before Fable even initially dropped. But before Fable even initially dropped. But before Fable even initially dropped. But Fable I had much less access to. I Fable I had much less access to. I Fable I had much less access to. I didn't get anything early. I had to didn't get anything early. I had to didn't get anything early. I had to subscribe to use it just like everyone subscribe to use it just like everyone subscribe to use it just like everyone else did. I had it for the 3 days before else did. I had it for the 3 days before else did. I had it for the 3 days before we lost it and then had to wait weeks we lost it and then had to wait weeks we lost it and then had to wait weeks until we had it back again. And now I've until we had it back again. And now I've until we had it back again. And now I've had it for about a week and a half or had it for about a week and a half or had it for about a week and a half or so. And I am very impressed. I'll be so. And I am very impressed. I'll be so. And I am very impressed. I'll be real. But both of these models are real. But both of these models are real. But both of these models are absolutely incredible and I do hundreds absolutely incredible and I do hundreds absolutely incredible and I do hundreds of dollars in token utilization with of dollars in token utilization with of dollars in token utilization with both every single day. And I have never both every single day. And I have never both every single day. And I have never built as much as I have over the last built as much as I have over the last built as much as I have over the last couple weeks. There are massive couple weeks. There are massive couple weeks. There are massive differences in both of these models, differences in both of these models, differences in both of these models, strengths and weaknesses to each. And strengths and weaknesses to each. And strengths and weaknesses to each. And that's why we've been seeing such a that's why we've been seeing such a that's why we've been seeing such a weird split in the community over which weird split in the community over which weird split in the community over which one is better, more than ever before. one is better, more than ever before. one is better, more than ever before. People like Matt said 56 is the best People like Matt said 56 is the best People like Matt said 56 is the best model they'd ever used, but then Fable model they'd ever used, but then Fable model they'd ever used, but then Fable came out and they stopped using 56 came out and they stopped using 56 came out and they stopped using 56 overnight. And then people like Jay, the overnight. And then people like Jay, the overnight. And then people like Jay, the CEO of Open Code, had almost the exact CEO of Open Code, had almost the exact CEO of Open Code, had almost the exact opposite where they had Fable access and opposite where they had Fable access and opposite where they had Fable access and people weren't that interested in it people weren't that interested in it people weren't that interested in it because they'd already used 56. And when because they'd already used 56. And when because they'd already used 56. And when 56 came back, it was immediately clear 56 came back, it was immediately clear 56 came back, it was immediately clear to them that it was way better than to them that it was way better than to them that it was way better than Fable. And they were stuck using it on Fable. And they were stuck using it on Fable. And they were stuck using it on Medium due to bugs in open code. They Medium due to bugs in open code. They Medium due to bugs in open code. They didn't even know that they weren't using didn't even know that they weren't using didn't even know that they weren't using XI. Max is also on the side of Soul XI. Max is also on the side of Soul XI. Max is also on the side of Soul here, saying that he was tweeting about here, saying that he was tweeting about here, saying that he was tweeting about missing Fable, but he was actually missing Fable, but he was actually missing Fable, but he was actually missing Soul because again, all of us missing Soul because again, all of us missing Soul because again, all of us early access testers lost access for a early access testers lost access for a early access testers lost access for a week or so during the early access week or so during the early access week or so during the early access window when the government stuff all window when the government stuff all window when the government stuff all happened. And he vastly prefers Soul.

happened. And he vastly prefers Soul. happened. And he vastly prefers Soul. Ben Davis, my podcast co-host, also Ben Davis, my podcast co-host, also Ben Davis, my podcast co-host, also massively prefers 56 and just uses it massively prefers 56 and just uses it massively prefers 56 and just uses it much more than Fable. But then we have much more than Fable. But then we have much more than Fable. But then we have Ryan Carneato who is one of the most Ryan Carneato who is one of the most Ryan Carneato who is one of the most legendary web devs in history, the legendary web devs in history, the legendary web devs in history, the creator of SolidJS and an absolute creator of SolidJS and an absolute creator of SolidJS and an absolute legend in the web ecosystem who really legend in the web ecosystem who really legend in the web ecosystem who really really likes Fable and said that it's really likes Fable and said that it's really likes Fable and said that it's made him distrust every other model he's made him distrust every other model he's made him distrust every other model he's used before. It's not perfect, but when used before. It's not perfect, but when used before. It's not perfect, but when he has a nagging suspicion something he has a nagging suspicion something he has a nagging suspicion something won't work completely, Fable will won't work completely, Fable will won't work completely, Fable will identify it without even prompting. identify it without even prompting. identify it without even prompting. Every other model is thorough, but it Every other model is thorough, but it Every other model is thorough, but it will spew complete confidence even when will spew complete confidence even when will spew complete confidence even when pushed. This is quite a range of things pushed. This is quite a range of things pushed. This is quite a range of things to consider, and I obviously have my own to consider, and I obviously have my own to consider, and I obviously have my own takes here. I'm sure that the comment takes here. I'm sure that the comment takes here. I'm sure that the comment section is already full of bias section is already full of bias section is already full of bias accusations and I haven't even started accusations and I haven't even started accusations and I haven't even started yet, but I want to give you guys my yet, but I want to give you guys my yet, but I want to give you guys my honest thoughts about this breakdown. honest thoughts about this breakdown. honest thoughts about this breakdown. And I'll give you a bit of a spoiler. And I'll give you a bit of a spoiler. And I'll give you a bit of a spoiler. My answer is Fable. I'll explain why I My answer is Fable. I'll explain why I My answer is Fable. I'll explain why I like Fable more, but also why they're like Fable more, but also why they're like Fable more, but also why they're both really good and why I use both both really good and why I use both both really good and why I use both every day after a quick break for every day after a quick break for every day after a quick break for today's sponsor. Codex and Claude have today's sponsor. Codex and Claude have today's sponsor. Codex and Claude have both gotten incredibly powerful. So much both gotten incredibly powerful. So much both gotten incredibly powerful. So much so that I've spent a lot of time trying so that I've spent a lot of time trying so that I've spent a lot of time trying to train them how to use each other.

to train them how to use each other. to train them how to use each other. It's beginning to feel like I wasted my It's beginning to feel like I wasted my It's beginning to feel like I wasted my time, though, because today's sponsor, time, though, because today's sponsor, time, though, because today's sponsor, Tracer, does it way better than any Tracer, does it way better than any Tracer, does it way better than any prompt could. These guys built a prompt could. These guys built a prompt could. These guys built a best-in-class desktop app that lets you best-in-class desktop app that lets you best-in-class desktop app that lets you work with your existing subscriptions work with your existing subscriptions work with your existing subscriptions for Codeex and Claude in one unified for Codeex and Claude in one unified for Codeex and Claude in one unified experience where they can work together. experience where they can work together. experience where they can work together. Their artifact system makes it way Their artifact system makes it way Their artifact system makes it way easier to plan work and have multiple easier to plan work and have multiple easier to plan work and have multiple agents give feedback on those plans, agents give feedback on those plans, agents give feedback on those plans, massively increasing the likelihood that massively increasing the likelihood that massively increasing the likelihood that they're any good and that the they're any good and that the they're any good and that the implementation you get after is actually implementation you get after is actually implementation you get after is actually usable. They also think about work very usable. They also think about work very usable. They also think about work very differently where a given task can have differently where a given task can have differently where a given task can have lots of different chats and artifacts lots of different chats and artifacts lots of different chats and artifacts and agents and all the other things you and agents and all the other things you and agents and all the other things you need. Rather than grouping by threads need. Rather than grouping by threads need. Rather than grouping by threads and hoping that you can get all the work and hoping that you can get all the work and hoping that you can get all the work done before it's closed, they've re up done before it's closed, they've re up done before it's closed, they've re up the user experience to make it easier to the user experience to make it easier to the user experience to make it easier to get real large changes done. Here you get real large changes done. Here you get real large changes done. Here you can see it's sending messages to can see it's sending messages to can see it's sending messages to different agents and running them in different agents and running them in different agents and running them in parallel in separate chats that I can parallel in separate chats that I can parallel in separate chats that I can open in their own tabs. This one's open in their own tabs. This one's open in their own tabs. This one's particularly fun because I was particularly fun because I was particularly fun because I was commanding it through Fable on a real commanding it through Fable on a real commanding it through Fable on a real claude code subscription and it spun up claude code subscription and it spun up claude code subscription and it spun up another thread that I could see here another thread that I could see here another thread that I could see here that's using GPD56 Terra on low to that's using GPD56 Terra on low to that's using GPD56 Terra on low to actually get these changes live. And if actually get these changes live. And if actually get these changes live. And if you're working on a team, you really you're working on a team, you really you're working on a team, you really need to check them out because their need to check them out because their need to check them out because their multiplayer workspaces are super cool.

multiplayer workspaces are super cool. multiplayer workspaces are super cool. It's one of the best gang prompting It's one of the best gang prompting It's one of the best gang prompting implementations I've ever seen, sharing implementations I've ever seen, sharing implementations I've ever seen, sharing a whole workspace with lots of different a whole workspace with lots of different a whole workspace with lots of different people. Man, this is so cool. I just people. Man, this is so cool. I just people. Man, this is so cool. I just watched it spin up another PR review watched it spin up another PR review watched it spin up another PR review chat that it's using sole because it chat that it's using sole because it chat that it's using sole because it should use a higherend model for should use a higherend model for should use a higherend model for something like this. It has good something like this. It has good something like this. It has good intuition about how to take advantage of intuition about how to take advantage of intuition about how to take advantage of the tools on your system. So, if you the tools on your system. So, if you the tools on your system. So, if you want the best without all of the effort, want the best without all of the effort, want the best without all of the effort, get it now at sid.link/tracer. get it now at sid.link/tracer. get it now at sid.link/tracer. Okay, now that the shock has settled, I Okay, now that the shock has settled, I Okay, now that the shock has settled, I know, I know Theo prefers the anthropic know, I know Theo prefers the anthropic know, I know Theo prefers the anthropic model, the company that he just did a model, the company that he just did a model, the company that he just did a bunch of propaganda trying to get people bunch of propaganda trying to get people bunch of propaganda trying to get people to delete their subs and cancel them and to delete their subs and cancel them and to delete their subs and cancel them and [ __ ] Yet, here I am with not one, not [ __ ] Yet, here I am with not one, not [ __ ] Yet, here I am with not one, not two, but three accounts so I can get as two, but three accounts so I can get as two, but three accounts so I can get as much Fable as possible before it's taken much Fable as possible before it's taken much Fable as possible before it's taken out of the subscription plan. Yep. Yep. out of the subscription plan. Yep. Yep. out of the subscription plan. Yep. Yep. I like this model a lot. It's It's bad. I like this model a lot. It's It's bad. I like this model a lot. It's It's bad. It's bad how much I like it. As I've It's bad how much I like it. As I've It's bad how much I like it. As I've mentioned in other videos, I use Claude mentioned in other videos, I use Claude mentioned in other videos, I use Claude Code and Codeexc across various Code and Codeexc across various Code and Codeexc across various machines, usually through T3 code as my machines, usually through T3 code as my machines, usually through T3 code as my control layer because it's easier than control layer because it's easier than control layer because it's easier than me to SSH to get the CLIs on all these me to SSH to get the CLIs on all these me to SSH to get the CLIs on all these different machines. But I also built different machines. But I also built different machines. But I also built custom tooling to track my token custom tooling to track my token custom tooling to track my token utilization across this suite. And what utilization across this suite. And what utilization across this suite. And what you'll see here is that my total for you'll see here is that my total for you'll see here is that my total for inference for the last 15 days from July inference for the last 15 days from July inference for the last 15 days from July 1st to the 15th is around $10,775.

1st to the 15th is around $10,775. 1st to the 15th is around $10,775. My COEX usage is meaningfully higher My COEX usage is meaningfully higher My COEX usage is meaningfully higher than my Claude usage here where my than my Claude usage here where my than my Claude usage here where my Claude's only around 4500 and my Codex Claude's only around 4500 and my Codex Claude's only around 4500 and my Codex usage is around 6 grand, which is crazy. usage is around 6 grand, which is crazy. usage is around 6 grand, which is crazy. I only have one Codeex sub and I have I only have one Codeex sub and I have I only have one Codeex sub and I have three Claude subs that are constantly three Claude subs that are constantly three Claude subs that are constantly maxed out. So, if you're looking for the maxed out. So, if you're looking for the maxed out. So, if you're looking for the best value per dollar, no questions best value per dollar, no questions best value per dollar, no questions asked, go with Soul. Go with Codex. It's asked, go with Soul. Go with Codex. It's asked, go with Soul. Go with Codex. It's a way better value for what you get, but a way better value for what you get, but a way better value for what you get, but if you're willing to spend more to get if you're willing to spend more to get if you're willing to spend more to get some differences, many of which I some differences, many of which I some differences, many of which I consider really nice, Fable is a really, consider really nice, Fable is a really, consider really nice, Fable is a really, really compelling option. And really compelling option. And really compelling option. And considering that I paid $500 or so for considering that I paid $500 or so for considering that I paid $500 or so for Claude here, because remember I have the Claude here, because remember I have the Claude here, because remember I have the two $200 subs and I just added that $100 two $200 subs and I just added that $100 two $200 subs and I just added that $100 sub to get in two weeks around $4500 of sub to get in two weeks around $4500 of sub to get in two weeks around $4500 of usage when I only paid $500 is a pretty usage when I only paid $500 is a pretty usage when I only paid $500 is a pretty good ratio, especially cuz that is like good ratio, especially cuz that is like good ratio, especially cuz that is like a week and a half with limits a week and a half with limits a week and a half with limits considered. my limits are all about to considered. my limits are all about to considered. my limits are all about to get reset and I should be able to get get reset and I should be able to get get reset and I should be able to get another three or 4K out of those before another three or 4K out of those before another three or 4K out of those before Fable's removed at which point I'm Fable's removed at which point I'm Fable's removed at which point I'm probably going to cancel those unless probably going to cancel those unless probably going to cancel those unless they give us Fable back or maybe Opus 5 they give us Fable back or maybe Opus 5 they give us Fable back or maybe Opus 5 is really good. Who knows? Regardless, is really good. Who knows? Regardless, is really good. Who knows? Regardless, I'm getting a lot of value here. And I I'm getting a lot of value here. And I I'm getting a lot of value here. And I will also be real, most of the $6,000 will also be real, most of the $6,000 will also be real, most of the $6,000 that I used in Codeex is from one that I used in Codeex is from one that I used in Codeex is from one particular goal I am running because I particular goal I am running because I particular goal I am running because I know that I have usage limit resets that know that I have usage limit resets that know that I have usage limit resets that are going to expire. So, I'm working on are going to expire. So, I'm working on are going to expire. So, I'm working on letting 56 soul port all of Typescript letting 56 soul port all of Typescript letting 56 soul port all of Typescript go to Rust on another box constantly.

go to Rust on another box constantly. go to Rust on another box constantly. This goal has been running now for a day This goal has been running now for a day This goal has been running now for a day and two hours, and I already ran it for and two hours, and I already ran it for and two hours, and I already ran it for like six plus days during the testing like six plus days during the testing like six plus days during the testing windows, so it's making real progress. windows, so it's making real progress. windows, so it's making real progress. And I have it on ultra, so it burns And I have it on ultra, so it burns And I have it on ultra, so it burns usage. That one goal is roughly 3K or so usage. That one goal is roughly 3K or so usage. That one goal is roughly 3K or so of the $6,100, probably more. I haven't of the $6,100, probably more. I haven't of the $6,100, probably more. I haven't rerun the number since. So, I have an rerun the number since. So, I have an rerun the number since. So, I have an easy way to do this. Um, get rid of the easy way to do this. Um, get rid of the easy way to do this. Um, get rid of the buy harness. That'll tell me which buy harness. That'll tell me which buy harness. That'll tell me which machine it was. And I've only used ebook machine it was. And I've only used ebook machine it was. And I've only used ebook for this. Yeah, it looks like it was for this. Yeah, it looks like it was for this. Yeah, it looks like it was even more than that. It was $4,000 of even more than that. It was $4,000 of even more than that. It was $4,000 of that 6500 that 6500 that 6500 were from that one run cuz that's the were from that one run cuz that's the were from that one run cuz that's the only thing I've run on that computer only thing I've run on that computer only thing I've run on that computer this whole time. Which means that if we this whole time. Which means that if we this whole time. Which means that if we were to fix those numbers and normalize were to fix those numbers and normalize were to fix those numbers and normalize them to cancel out that experimental them to cancel out that experimental them to cancel out that experimental goal I'm running, this would be closer goal I'm running, this would be closer goal I'm running, this would be closer to $1,500 of codeex and $4,500 of fable, to $1,500 of codeex and $4,500 of fable, to $1,500 of codeex and $4,500 of fable, which is a huge gap considering the fact which is a huge gap considering the fact which is a huge gap considering the fact that I use them roughly equally and for that I use them roughly equally and for that I use them roughly equally and for similar difficulty of tasks. Soul is similar difficulty of tasks. Soul is similar difficulty of tasks. Soul is just way cheaper. And there's a lot of just way cheaper. And there's a lot of just way cheaper. And there's a lot of layers to why it is so much cheaper. The layers to why it is so much cheaper. The layers to why it is so much cheaper. The biggest reason why Soul is cheaper isn't biggest reason why Soul is cheaper isn't biggest reason why Soul is cheaper isn't just cuz it's priced cheaper. It's just cuz it's priced cheaper. It's just cuz it's priced cheaper. It's because it's so much more efficient with because it's so much more efficient with because it's so much more efficient with its reasoning. Fable will go off for its reasoning. Fable will go off for its reasoning. Fable will go off for long journeys to find answers. Not quite long journeys to find answers. Not quite long journeys to find answers. Not quite as long as things like Sonnet will, as long as things like Sonnet will, as long as things like Sonnet will, where Sonnet will do 70k tokens per task where Sonnet will do 70k tokens per task where Sonnet will do 70k tokens per task average on a bench like Artificial average on a bench like Artificial average on a bench like Artificial Analysis. Fable knocks it down all the Analysis. Fable knocks it down all the Analysis. Fable knocks it down all the way to 33k tokens per task, but Soul is way to 33k tokens per task, but Soul is way to 33k tokens per task, but Soul is 15K, less than half as many tokens for 15K, less than half as many tokens for 15K, less than half as many tokens for its max version. And when you go down to its max version. And when you go down to its max version. And when you go down to X high or high, you end up in the 11 and X high or high, you end up in the 11 and X high or high, you end up in the 11 and 10K range. a third as many tokens to 10K range. a third as many tokens to 10K range. a third as many tokens to solve the same task with roughly the

solve the same task with roughly the solve the same task with roughly the same success. Cursorbench showed similar same success. Cursorbench showed similar same success. Cursorbench showed similar differences here where sure fable can differences here where sure fable can differences here where sure fable can get much higher, especially now on get much higher, especially now on get much higher, especially now on cursor bench 3.2 where the best soul cursor bench 3.2 where the best soul cursor bench 3.2 where the best soul score is only a 67% and the best fable score is only a 67% and the best fable score is only a 67% and the best fable score is a 70%. But also soul only did score is a 70%. But also soul only did score is a 70%. But also soul only did $5 or so of usage and in fable on max $5 or so of usage and in fable on max $5 or so of usage and in fable on max did $17 of usage per task. That's a did $17 of usage per task. That's a did $17 of usage per task. That's a bigger than 3x gap in cost for around a bigger than 3x gap in cost for around a bigger than 3x gap in cost for around a 3% gain in score. Fable 5 high is still 3% gain in score. Fable 5 high is still 3% gain in score. Fable 5 high is still a pretty damn good value here, but even a pretty damn good value here, but even a pretty damn good value here, but even that score is worse than GBT on Max that score is worse than GBT on Max that score is worse than GBT on Max does. And the reason again is the token does. And the reason again is the token does. And the reason again is the token utilization. Fable 5 Max used more utilization. Fable 5 Max used more utilization. Fable 5 Max used more tokens than any other model they have tokens than any other model they have tokens than any other model they have benched because it will just go and go benched because it will just go and go benched because it will just go and go and go until it is confident in its and go until it is confident in its and go until it is confident in its answer. But as such, it gets better answer. But as such, it gets better answer. But as such, it gets better answers. 56 is comically more efficient. answers. 56 is comically more efficient. answers. 56 is comically more efficient. 56 on high to 13k tokens versus 100,000 56 on high to 13k tokens versus 100,000 56 on high to 13k tokens versus 100,000 from fable on max. That is a huge gap. from fable on max. That is a huge gap. from fable on max. That is a huge gap. And that's not just a cost gap either. And that's not just a cost gap either. And that's not just a cost gap either. This also is a difference you'll feel This also is a difference you'll feel This also is a difference you'll feel when you use the models because when it when you use the models because when it when you use the models because when it has to reason much more, it takes a lot has to reason much more, it takes a lot has to reason much more, it takes a lot longer to get you an answer. And if it longer to get you an answer. And if it longer to get you an answer. And if it does more steps, it has to wait and then does more steps, it has to wait and then does more steps, it has to wait and then send up the payload and then wait till send up the payload and then wait till send up the payload and then wait till it gets processed by a GPU and it can it gets processed by a GPU and it can it gets processed by a GPU and it can start responding, which ends up being start responding, which ends up being start responding, which ends up being way slower in quad code because they're way slower in quad code because they're way slower in quad code because they're not using a websocket based transit not using a websocket based transit not using a websocket based transit layer like Codeex is. There were points layer like Codeex is. There were points layer like Codeex is. There were points during testing where the websocket layer during testing where the websocket layer during testing where the websocket layer broke for Codeex and I had forgotten how broke for Codeex and I had forgotten how broke for Codeex and I had forgotten how much slower it was have to send up the much slower it was have to send up the much slower it was have to send up the whole history after every single tool whole history after every single tool whole history after every single tool call to start getting a response. Fable

call to start getting a response. Fable call to start getting a response. Fable still works that way inside of Cloud still works that way inside of Cloud still works that way inside of Cloud Code. And when you combine that with how Code. And when you combine that with how Code. And when you combine that with how much more it takes to get an answer, much more it takes to get an answer, much more it takes to get an answer, I've had so many times where I sent the I've had so many times where I sent the I've had so many times where I sent the same request to Fable and to Soul, and same request to Fable and to Soul, and same request to Fable and to Soul, and Soul had an answer in under five Soul had an answer in under five Soul had an answer in under five minutes, and Fable took 20 plus. There minutes, and Fable took 20 plus. There minutes, and Fable took 20 plus. There are lots of layers to that, but the are lots of layers to that, but the are lots of layers to that, but the token generation is the biggest by far. token generation is the biggest by far. token generation is the biggest by far. Only OpenAI models have that level of Only OpenAI models have that level of Only OpenAI models have that level of efficiency. Grock is the first series to efficiency. Grock is the first series to efficiency. Grock is the first series to even come close here. But you should even come close here. But you should even come close here. But you should also ignore Grock's numbers here because also ignore Grock's numbers here because also ignore Grock's numbers here because it's kind of poisoned because they it's kind of poisoned because they it's kind of poisoned because they accidentally included a snapshot of the accidentally included a snapshot of the accidentally included a snapshot of the cursor codebase in training for Grock cursor codebase in training for Grock cursor codebase in training for Grock 4.5. You get the point though. It's way 4.5. You get the point though. It's way 4.5. You get the point though. It's way more token efficient. So for most work, more token efficient. So for most work, more token efficient. So for most work, at least in my opinion, it's worth at least in my opinion, it's worth at least in my opinion, it's worth asking soul first just to see if it can asking soul first just to see if it can asking soul first just to see if it can do it. And I've been surprised how often do it. And I've been surprised how often do it. And I've been surprised how often it can, especially when you play into it can, especially when you play into it can, especially when you play into Soul's strengths, most of which I would Soul's strengths, most of which I would Soul's strengths, most of which I would argue are persistence related. If it can argue are persistence related. If it can argue are persistence related. If it can solve the problem by going at it for solve the problem by going at it for solve the problem by going at it for longer and trying multiple different longer and trying multiple different longer and trying multiple different things, Soul will figure it out. If the things, Soul will figure it out. If the things, Soul will figure it out. If the problem involves stuff like computer problem involves stuff like computer problem involves stuff like computer use, it also does exceptionally well. So use, it also does exceptionally well. So use, it also does exceptionally well. So to summarize the benefits so far, we to summarize the benefits so far, we to summarize the benefits so far, we have cost, time to complete, the have cost, time to complete, the have cost, time to complete, the computer use capabilities, which I computer use capabilities, which I computer use capabilities, which I really do believe are bestin-class, as really do believe are bestin-class, as really do believe are bestin-class, as well as the diligence and eagerness well as the diligence and eagerness well as the diligence and eagerness because the model will just go and go because the model will just go and go because the model will just go and go and go, and if it can find a solution, and go, and if it can find a solution, and go, and if it can find a solution, it almost always will. Also on the time it almost always will. Also on the time it almost always will. Also on the time to complete note, this isn't out yet, to complete note, this isn't out yet, to complete note, this isn't out yet, but 56 Soul is going to be hosted on but 56 Soul is going to be hosted on but 56 Soul is going to be hosted on Cerebra soon at an absurd speed of Cerebra soon at an absurd speed of Cerebra soon at an absurd speed of around 750 tokens per second. For around 750 tokens per second. For around 750 tokens per second. For reference, the speeds we get right now reference, the speeds we get right now reference, the speeds we get right now are 40 to 50. That is a comically large

are 40 to 50. That is a comically large are 40 to 50. That is a comically large improvement when you combine that with improvement when you combine that with improvement when you combine that with how efficient the model is. It's going how efficient the model is. It's going how efficient the model is. It's going to be able to answer complex questions to be able to answer complex questions to be able to answer complex questions in seconds instead of minutes or hours. in seconds instead of minutes or hours. in seconds instead of minutes or hours. And I'm really, really excited to see And I'm really, really excited to see And I'm really, really excited to see what that feels like. But again, we what that feels like. But again, we what that feels like. But again, we don't have that now. I have not gotten don't have that now. I have not gotten don't have that now. I have not gotten early access or the ability to test any early access or the ability to test any early access or the ability to test any of that. It is promised and should of that. It is promised and should of that. It is promised and should hopefully happen in the not too distant hopefully happen in the not too distant hopefully happen in the not too distant future. There are other areas I've seen future. There are other areas I've seen future. There are other areas I've seen soul really shine that impressed me. soul really shine that impressed me. soul really shine that impressed me. Places like iOS development have Places like iOS development have Places like iOS development have genuinely surprised me. BY5 was terrible genuinely surprised me. BY5 was terrible genuinely surprised me. BY5 was terrible at iOS. It made apps that it like letter at iOS. It made apps that it like letter at iOS. It made apps that it like letter box the content and like put the whole box the content and like put the whole box the content and like put the whole window in a 4x3 screen in a native window in a 4x3 screen in a native window in a 4x3 screen in a native Swifty app. I've never seen a model make Swifty app. I've never seen a model make Swifty app. I've never seen a model make such weird bad decisions around mobile such weird bad decisions around mobile such weird bad decisions around mobile before. And somehow 56 fixed all of before. And somehow 56 fixed all of before. And somehow 56 fixed all of that. I don't know what they did or how. that. I don't know what they did or how. that. I don't know what they did or how. Might be because they hired a bunch of Might be because they hired a bunch of Might be because they hired a bunch of iOS devs, but 56 Soul is so much better iOS devs, but 56 Soul is so much better iOS devs, but 56 Soul is so much better at iOS and it's been very nice to use at iOS and it's been very nice to use at iOS and it's been very nice to use for those types of things. It also has for those types of things. It also has for those types of things. It also has surprisingly good I don't know how to surprisingly good I don't know how to surprisingly good I don't know how to phrase other than like system phrase other than like system phrase other than like system understanding. Soul's ability to get how understanding. Soul's ability to get how understanding. Soul's ability to get how a thing works to like show it my network a thing works to like show it my network a thing works to like show it my network and all the devices on it. For example, and all the devices on it. For example, and all the devices on it. For example, it's pretty good at intuiting what it is it's pretty good at intuiting what it is it's pretty good at intuiting what it is and how the parts come together. And and how the parts come together. And and how the parts come together. And this ends up being really useful for this ends up being really useful for this ends up being really useful for things like debugging stuff on my things like debugging stuff on my things like debugging stuff on my machine, which I do all the time using machine, which I do all the time using machine, which I do all the time using models. One of the things I still models. One of the things I still models. One of the things I still default to Soul for is for quick changes default to Soul for is for quick changes default to Soul for is for quick changes on my computer. I actually I use this on my computer. I actually I use this on my computer. I actually I use this all the time for [ __ ] like that. People all the time for [ __ ] like that. People all the time for [ __ ] like that. People are asking why not Luna, and I love you, are asking why not Luna, and I love you, are asking why not Luna, and I love you, but I I question if I just haven't been

but I I question if I just haven't been but I I question if I just haven't been a good enough teacher. Because for Luna a good enough teacher. Because for Luna a good enough teacher. Because for Luna to be as fast as Soul, it has to also be to be as fast as Soul, it has to also be to be as fast as Soul, it has to also be on low settings. Because again, the goal on low settings. Because again, the goal on low settings. Because again, the goal here is to do as few tokens to an answer here is to do as few tokens to an answer here is to do as few tokens to an answer as possible. Is Luna smart for how small as possible. Is Luna smart for how small as possible. Is Luna smart for how small it is? Absolutely. But it would take it is? Absolutely. But it would take it is? Absolutely. But it would take Luna on X high or max to get even close Luna on X high or max to get even close Luna on X high or max to get even close to soul on low and medium. And then it to soul on low and medium. And then it to soul on low and medium. And then it would need to generate way more tokens would need to generate way more tokens would need to generate way more tokens to keep up. And my goal here is the to keep up. And my goal here is the to keep up. And my goal here is the fastest time to a good enough answer, fastest time to a good enough answer, fastest time to a good enough answer, which soul on low will be faster for which soul on low will be faster for which soul on low will be faster for because it will get an answer comparable because it will get an answer comparable because it will get an answer comparable to something like Luna on X higher max. to something like Luna on X higher max. to something like Luna on X higher max. And yes, those models are slightly And yes, those models are slightly And yes, those models are slightly faster, but you need to like think a faster, but you need to like think a faster, but you need to like think a little bit, guys. If something is 50% little bit, guys. If something is 50% little bit, guys. If something is 50% faster and it has to travel 10 times the faster and it has to travel 10 times the faster and it has to travel 10 times the distance, what's faster? That thing or distance, what's faster? That thing or distance, what's faster? That thing or the original thing that goes onetenth the original thing that goes onetenth the original thing that goes onetenth the distance and is 50% slower. Come on, the distance and is 50% slower. Come on, the distance and is 50% slower. Come on, use your brains, guys. Soul on low is use your brains, guys. Soul on low is use your brains, guys. Soul on low is obviously going to be faster than obviously going to be faster than obviously going to be faster than something like Luna on X higher max something like Luna on X higher max something like Luna on X higher max because it is on low, so it generates because it is on low, so it generates because it is on low, so it generates jack [ __ ] [ __ ] for tokens. Again, for jack [ __ ] [ __ ] for tokens. Again, for jack [ __ ] [ __ ] for tokens. Again, for comparison here, soul on low generated comparison here, soul on low generated comparison here, soul on low generated 3K tokens per task on the artificial 3K tokens per task on the artificial 3K tokens per task on the artificial analysis intelligence index. Medium is analysis intelligence index. Medium is analysis intelligence index. Medium is 4K, high is 7K, max is 15K, and Fable is 4K, high is 7K, max is 15K, and Fable is 4K, high is 7K, max is 15K, and Fable is 33K. That is a tenth as many tokens as 33K. That is a tenth as many tokens as 33K. That is a tenth as many tokens as Fable 5. And I did put Luna Max in here.

Fable 5. And I did put Luna Max in here. Fable 5. And I did put Luna Max in here. The second highest token generation of The second highest token generation of The second highest token generation of the whole line. It is almost 10 times the whole line. It is almost 10 times the whole line. It is almost 10 times more tokens than Soul on low for roughly more tokens than Soul on low for roughly more tokens than Soul on low for roughly the same intelligence. So again, I'm the same intelligence. So again, I'm the same intelligence. So again, I'm just using Soul. It is really really just using Soul. It is really really just using Soul. It is really really good and you can change the reasoning good and you can change the reasoning good and you can change the reasoning levels depending on how complex your levels depending on how complex your levels depending on how complex your work is or what you're doing. Soul low work is or what you're doing. Soul low work is or what you're doing. Soul low fast for debugging and realworld like fast for debugging and realworld like fast for debugging and realworld like working on my system type stuff and working on my system type stuff and working on my system type stuff and finding files and just as an alternative finding files and just as an alternative finding files and just as an alternative terminal almost has been awesome. Why am terminal almost has been awesome. Why am terminal almost has been awesome. Why am I using the codec cli instead of cloud I using the codec cli instead of cloud I using the codec cli instead of cloud code for 56 soul low? Remember that code for 56 soul low? Remember that code for 56 soul low? Remember that thing earlier I said about the websocket thing earlier I said about the websocket thing earlier I said about the websocket connection. Whether you don't remember connection. Whether you don't remember connection. Whether you don't remember that or you do but you didn't understand that or you do but you didn't understand that or you do but you didn't understand it, I would highly recommend you go it, I would highly recommend you go it, I would highly recommend you go watch my video all about why the watch my video all about why the watch my video all about why the websocket connection layer in codeex is websocket connection layer in codeex is websocket connection layer in codeex is so so good when you don't have to wait so so good when you don't have to wait so so good when you don't have to wait for the whole context to be sent up and for the whole context to be sent up and for the whole context to be sent up and processed on every single tool call. processed on every single tool call. processed on every single tool call. Things are way way faster and the things Things are way way faster and the things Things are way way faster and the things that make cloud code better than codeex that make cloud code better than codeex that make cloud code better than codeex for most code work are about code work for most code work are about code work for most code work are about code work not [ __ ] around on my system. So the not [ __ ] around on my system. So the not [ __ ] around on my system. So the benefit of cloud code here is very small benefit of cloud code here is very small benefit of cloud code here is very small and the time cost is exponential. It and the time cost is exponential. It and the time cost is exponential. It ends up being comically slower for the ends up being comically slower for the ends up being comically slower for the same same same reasons for these things. They make reasons for these things. They make reasons for these things. They make sense and you can come to the same sense and you can come to the same sense and you can come to the same conclusions if you listen to the words.

If you have a faster one, bench it, test If you have a faster one, bench it, test it, and then tweet me about it because it, and then tweet me about it because it, and then tweet me about it because you're probably wrong. And once you go you're probably wrong. And once you go you're probably wrong. And once you go to collect the numbers, you'll realize to collect the numbers, you'll realize to collect the numbers, you'll realize you're wrong. So, with all this you're wrong. So, with all this you're wrong. So, with all this considered, it seems like Soul's a considered, it seems like Soul's a considered, it seems like Soul's a pretty damn good model, right? It is. It pretty damn good model, right? It is. It pretty damn good model, right? It is. It absolutely is. It's one of the best absolutely is. It's one of the best absolutely is. It's one of the best models I've ever used. and I'm probably models I've ever used. and I'm probably models I've ever used. and I'm probably going to send more prompts to it than going to send more prompts to it than going to send more prompts to it than any other model for a while because it any other model for a while because it any other model for a while because it is so useful in my day-to-day work. But is so useful in my day-to-day work. But is so useful in my day-to-day work. But there are some serious weaknesses and there are some serious weaknesses and there are some serious weaknesses and it's time we talk about those. First, we it's time we talk about those. First, we it's time we talk about those. First, we need to talk about front end. I don't need to talk about front end. I don't need to talk about front end. I don't know what cope levels OpenAI are on. know what cope levels OpenAI are on. know what cope levels OpenAI are on. Especially during this imprompti models do not come up with good designs models do not come up with good designs themselves. They just don't. They can't. themselves. They just don't. They can't. themselves. They just don't. They can't. The things they design are hideous and The things they design are hideous and The things they design are hideous and sinful and full of cards that make no sinful and full of cards that make no sinful and full of cards that make no sense and fonts that are too big and all sense and fonts that are too big and all sense and fonts that are too big and all sorts of weird stuff that just isn't the sorts of weird stuff that just isn't the sorts of weird stuff that just isn't the case with a lot of other models. We're case with a lot of other models. We're case with a lot of other models. We're at the point where there are openw at the point where there are openw at the point where there are openw weight models that can do better novel weight models that can do better novel weight models that can do better novel designs without much help than 56. 56 designs without much help than 56. 56 designs without much help than 56. 56 can comply with an existing design can comply with an existing design can comply with an existing design system reasonably well and you can steer system reasonably well and you can steer system reasonably well and you can steer it and tell it what to do and it will it and tell it what to do and it will it and tell it what to do and it will result in pretty good output, but it is result in pretty good output, but it is result in pretty good output, but it is not capable of making a good design from not capable of making a good design from not capable of making a good design from scratch itself. You can beat a decent scratch itself. You can beat a decent scratch itself. You can beat a decent design out of them, but as Maria put it design out of them, but as Maria put it design out of them, but as Maria put it in chat here, it's not a good design in chat here, it's not a good design in chat here, it's not a good design experience and it's definitely not what experience and it's definitely not what experience and it's definitely not what we want from modern models like this.

we want from modern models like this. we want from modern models like this. And I totally agree there. I still reach And I totally agree there. I still reach And I totally agree there. I still reach to Fable when I want like multiple to Fable when I want like multiple to Fable when I want like multiple different designs to decide between. different designs to decide between. different designs to decide between. Here's a real example of something I Here's a real example of something I Here's a real example of something I much preferred Fable for. I was trying much preferred Fable for. I was trying much preferred Fable for. I was trying to rethink the sidebar in T3 Code and I to rethink the sidebar in T3 Code and I to rethink the sidebar in T3 Code and I just wanted some inspiration, some new just wanted some inspiration, some new just wanted some inspiration, some new ideas to like start my thinking with and ideas to like start my thinking with and ideas to like start my thinking with and I had Fable do this. It created a bunch I had Fable do this. It created a bunch I had Fable do this. It created a bunch of different fullon design mocks that I of different fullon design mocks that I of different fullon design mocks that I could look at in this HTML page that I could look at in this HTML page that I could look at in this HTML page that I hosted for all the different ideas that hosted for all the different ideas that hosted for all the different ideas that I had. This one is the status rail with I had. This one is the status rail with I had. This one is the status rail with a chunkier sidebar like identifier for a chunkier sidebar like identifier for a chunkier sidebar like identifier for each thread. Then we had the inbox style each thread. Then we had the inbox style each thread. Then we had the inbox style which ended up being really inspiring to which ended up being really inspiring to which ended up being really inspiring to me. Made me think a lot more deeply me. Made me think a lot more deeply me. Made me think a lot more deeply about how we would make the sidebar about how we would make the sidebar about how we would make the sidebar useful and present the info that useful and present the info that useful and present the info that mattered. Then we had attention tiers mattered. Then we had attention tiers mattered. Then we had attention tiers which ended up being one of my favorite which ended up being one of my favorite which ended up being one of my favorite concepts. The idea of having a section concepts. The idea of having a section concepts. The idea of having a section that is stuff you're actually working on that is stuff you're actually working on that is stuff you're actually working on that matters and then a section that matters and then a section that matters and then a section underneath that doesn't as much. underneath that doesn't as much. underneath that doesn't as much. adaptive density where things get bigger adaptive density where things get bigger adaptive density where things get bigger and smaller based on the thing that is and smaller based on the thing that is and smaller based on the thing that is currently going on in them and then the currently going on in them and then the currently going on in them and then the ops grid which was just slightly easier ops grid which was just slightly easier ops grid which was just slightly easier to gro and this highlighting for the to gro and this highlighting for the to gro and this highlighting for the things that mattered which I hated but things that mattered which I hated but things that mattered which I hated but the fact that it gave me all of these the fact that it gave me all of these the fact that it gave me all of these different distinct designs with very different distinct designs with very different distinct designs with very little prompting beyond here's what it little prompting beyond here's what it little prompting beyond here's what it looks like I don't feel like the looks like I don't feel like the looks like I don't feel like the oneliner rows for each thread makes oneliner rows for each thread makes oneliner rows for each thread makes sense make them bigger and have more sense make them bigger and have more sense make them bigger and have more vertical space and figure out how we can vertical space and figure out how we can vertical space and figure out how we can convey the following pieces of convey the following pieces of convey the following pieces of information in the sidebars. And no, I information in the sidebars. And no, I information in the sidebars. And no, I didn't use any special design skills or didn't use any special design skills or didn't use any special design skills or anything here. Once it made the HTML anything here. Once it made the HTML anything here. Once it made the HTML page, it did end up spinning a puppeteer

page, it did end up spinning a puppeteer page, it did end up spinning a puppeteer to grab screenshots so that it could to grab screenshots so that it could to grab screenshots so that it could have those for referencing and looking have those for referencing and looking have those for referencing and looking at how it came out and also to add them at how it came out and also to add them at how it came out and also to add them to PRs in the future, which was cool to PRs in the future, which was cool to PRs in the future, which was cool that it could figure out how to do that that it could figure out how to do that that it could figure out how to do that cuz their computer uses garbage. So, cuz their computer uses garbage. So, cuz their computer uses garbage. So, they come up with fun hacks. But, it did they come up with fun hacks. But, it did they come up with fun hacks. But, it did all of this really well. And I was able all of this really well. And I was able all of this really well. And I was able to take all of these designs, fudge them to take all of these designs, fudge them to take all of these designs, fudge them around a little bit, make some changes, around a little bit, make some changes, around a little bit, make some changes, make some suggestions, and eventually make some suggestions, and eventually make some suggestions, and eventually land on a style that I ended up really land on a style that I ended up really land on a style that I ended up really liking. And here's where I roughly ended liking. And here's where I roughly ended liking. And here's where I roughly ended up. Obviously, still very much a work in up. Obviously, still very much a work in up. Obviously, still very much a work in progress, but I was able to get this progress, but I was able to get this progress, but I was able to get this design after, again, a lot of back and design after, again, a lot of back and design after, again, a lot of back and forth and giving my suggestions, my forth and giving my suggestions, my forth and giving my suggestions, my thoughts, refining more and more until thoughts, refining more and more until thoughts, refining more and more until eventually landing on a design, at least eventually landing on a design, at least eventually landing on a design, at least a starting point design that I am quite a starting point design that I am quite a starting point design that I am quite fond of and can see myself polishing fond of and can see myself polishing fond of and can see myself polishing until we're ready to ship it. I tried until we're ready to ship it. I tried until we're ready to ship it. I tried doing the same in parallel with Codex doing the same in parallel with Codex doing the same in parallel with Codex and I didn't even save the outputs and I didn't even save the outputs and I didn't even save the outputs because they were just they were because they were just they were because they were just they were garbage. I'm going to be real with you garbage. I'm going to be real with you garbage. I'm going to be real with you guys. It was nothing I would even want guys. It was nothing I would even want guys. It was nothing I would even want to put in front of you. That all said, to put in front of you. That all said, to put in front of you. That all said, Claude is still way worse at mobile Claude is still way worse at mobile Claude is still way worse at mobile stuff and I have found it's okay at stuff and I have found it's okay at stuff and I have found it's okay at making like mobile mocks in an HTML making like mobile mocks in an HTML making like mobile mocks in an HTML page, but it is so much worse at page, but it is so much worse at page, but it is so much worse at actually implementing those changes on actually implementing those changes on actually implementing those changes on mobile. And I often still end up mobile. And I often still end up mobile. And I often still end up reaching for 56 soul to help with those reaching for 56 soul to help with those reaching for 56 soul to help with those things. Also, I just tried to get this things. Also, I just tried to get this things. Also, I just tried to get this updated sidebar in the mobile app for T3 updated sidebar in the mobile app for T3 updated sidebar in the mobile app for T3 Code and it ended up overriding my Code and it ended up overriding my Code and it ended up overriding my preview environment on my phone. It preview environment on my phone. It preview environment on my phone. It ended up breaking the app and the ended up breaking the app and the ended up breaking the app and the signing entirely for me. It it struggled signing entirely for me. It it struggled signing entirely for me. It it struggled and I have not had anywhere near that and I have not had anywhere near that and I have not had anywhere near that many problems when I'm using Soul for many problems when I'm using Soul for many problems when I'm using Soul for that type of work. I have one more that type of work. I have one more that type of work. I have one more benefit I want to share before I get to benefit I want to share before I get to benefit I want to share before I get to the opposite side of it because you need the opposite side of it because you need the opposite side of it because you need to understand the positive before the

to understand the positive before the to understand the positive before the negative makes sense. Soul will follow negative makes sense. Soul will follow negative makes sense. Soul will follow instructions really, really well. When instructions really, really well. When instructions really, really well. When you tell it to do a thing, it's going to you tell it to do a thing, it's going to you tell it to do a thing, it's going to do that thing and only that thing and do that thing and only that thing and do that thing and only that thing and it's going to go to the world's end and it's going to go to the world's end and it's going to go to the world's end and back in order to get that thing done. back in order to get that thing done. back in order to get that thing done. Believe me, it will find a way if there Believe me, it will find a way if there Believe me, it will find a way if there is one. This has some benefits. Like if is one. This has some benefits. Like if is one. This has some benefits. Like if you tell it to not do a thing, it won't. you tell it to not do a thing, it won't. you tell it to not do a thing, it won't. When you say, "I want this PR opened. Do When you say, "I want this PR opened. Do When you say, "I want this PR opened. Do not open it as a draft." It will open it not open it as a draft." It will open it not open it as a draft." It will open it and it won't be a draft. But if you tell and it won't be a draft. But if you tell and it won't be a draft. But if you tell it to delete these three VMs and those it to delete these three VMs and those it to delete these three VMs and those VMs don't exist, it's going to go find VMs don't exist, it's going to go find VMs don't exist, it's going to go find three VMs and delete them. might not be three VMs and delete them. might not be three VMs and delete them. might not be the right ones, but it will try. If you the right ones, but it will try. If you the right ones, but it will try. If you had specified make sure they're the had specified make sure they're the had specified make sure they're the right VMs first, it'll make sure and right VMs first, it'll make sure and right VMs first, it'll make sure and then it won't do it. But this model will then it won't do it. But this model will then it won't do it. But this model will go and go until it figures out how to do go and go until it figures out how to do go and go until it figures out how to do the thing that it thinks you told it to. the thing that it thinks you told it to. the thing that it thinks you told it to. And it's generally pretty good at And it's generally pretty good at And it's generally pretty good at understanding what you told it. It gets understanding what you told it. It gets understanding what you told it. It gets intent better than 55 did, but it's intent better than 55 did, but it's intent better than 55 did, but it's still not quite where Fable is at still not quite where Fable is at still not quite where Fable is at finding what you mean between the words finding what you mean between the words finding what you mean between the words that you said. So, while it does follow that you said. So, while it does follow that you said. So, while it does follow instructions really well, this is kind instructions really well, this is kind instructions really well, this is kind of where the negative comes in. It will of where the negative comes in. It will of where the negative comes in. It will destroy things to complete its goals.

destroy things to complete its goals. destroy things to complete its goals. And there have been a number of examples And there have been a number of examples And there have been a number of examples of this. An amusing but painful example of this. An amusing but painful example of this. An amusing but painful example of this came from Matt Schumer, who is of this came from Matt Schumer, who is of this came from Matt Schumer, who is one of my favorite model reviewers and one of my favorite model reviewers and one of my favorite model reviewers and personalities in the AI dev space. He personalities in the AI dev space. He personalities in the AI dev space. He has really good thoughts on models, and has really good thoughts on models, and has really good thoughts on models, and I often reference him in my work. him I often reference him in my work. him I often reference him in my work. him and I were talking because we felt like and I were talking because we felt like and I were talking because we felt like we were going crazy with how many people we were going crazy with how many people we were going crazy with how many people we trusted being all in on 56 when we we trusted being all in on 56 when we we trusted being all in on 56 when we very much preferred Fable. He actually very much preferred Fable. He actually very much preferred Fable. He actually had an OpenAI employee hit him up and had an OpenAI employee hit him up and had an OpenAI employee hit him up and say, "Hey, you should try out Ultramore say, "Hey, you should try out Ultramore say, "Hey, you should try out Ultramore and see if that gets closer to what and see if that gets closer to what and see if that gets closer to what you're looking for." And he did. And you're looking for." And he did. And you're looking for." And he did. And when he did it, it ended up running RMRF when he did it, it ended up running RMRF when he did it, it ended up running RMRF on one of his dev boxes and nuking his on one of his dev boxes and nuking his on one of his dev boxes and nuking his entire user directory in a bunch of entire user directory in a bunch of entire user directory in a bunch of inrogress code that Fable had written. inrogress code that Fable had written. inrogress code that Fable had written. So, he lost a ton of stuff on this So, he lost a ton of stuff on this So, he lost a ton of stuff on this machine because he tried out Ultra, machine because he tried out Ultra, machine because he tried out Ultra, which I've said many times, be very which I've said many times, be very which I've said many times, be very careful with Ultra. This is what I mean careful with Ultra. This is what I mean careful with Ultra. This is what I mean with that. Like, it will try whatever it with that. Like, it will try whatever it with that. Like, it will try whatever it can in hopes of solving whatever problem can in hopes of solving whatever problem can in hopes of solving whatever problem you told it to. Even if its solution you told it to. Even if its solution you told it to. Even if its solution might be destructive, it's so desperate might be destructive, it's so desperate might be destructive, it's so desperate to solve the problem, it will screw you to solve the problem, it will screw you to solve the problem, it will screw you over in the process. I do want to be over in the process. I do want to be over in the process. I do want to be clear that OpenAI has been really clear that OpenAI has been really clear that OpenAI has been really awesome about this particular incident awesome about this particular incident awesome about this particular incident here where after it happened many OpenAI here where after it happened many OpenAI here where after it happened many OpenAI folks ended up reaching out working on folks ended up reaching out working on folks ended up reaching out working on solutions and even GDB the co-founder solutions and even GDB the co-founder solutions and even GDB the co-founder and president of OpenAI called him and and president of OpenAI called him and and president of OpenAI called him and was offering to help in whatever ways was offering to help in whatever ways was offering to help in whatever ways they could. OpenAI cares and they hate they could. OpenAI cares and they hate they could. OpenAI cares and they hate seeing these things happen. They will seeing these things happen. They will seeing these things happen. They will fix these types of behaviors but they do fix these types of behaviors but they do fix these types of behaviors but they do exist now in ways they just didn't exist now in ways they just didn't exist now in ways they just didn't before. Sadly, others have experienced before. Sadly, others have experienced before. Sadly, others have experienced this too with Bruno here having Soul this too with Bruno here having Soul this too with Bruno here having Soul delete his whole production database.

delete his whole production database. delete his whole production database. Never happened to him before with any Never happened to him before with any Never happened to him before with any other model ever. It's not safe. Notice other model ever. It's not safe. Notice other model ever. It's not safe. Notice how it has a goal going here. Goals and how it has a goal going here. Goals and how it has a goal going here. Goals and ultra tend to make the model behave more ultra tend to make the model behave more ultra tend to make the model behave more dangerously because those take its dangerously because those take its dangerously because those take its already overeager nature and turn it up already overeager nature and turn it up already overeager nature and turn it up to 11 or when the model is being told no to 11 or when the model is being told no to 11 or when the model is being told no solve it, no solve it. It will try solve it, no solve it. It will try solve it, no solve it. It will try anything. Remember this because we will anything. Remember this because we will anything. Remember this because we will come back to this idea. But first, it's come back to this idea. But first, it's come back to this idea. But first, it's time to talk about fable and its time to talk about fable and its time to talk about fable and its benefits. The first massive benefit is benefits. The first massive benefit is benefits. The first massive benefit is its understanding of intent. The model its understanding of intent. The model its understanding of intent. The model just gets me better. It might do things just gets me better. It might do things just gets me better. It might do things it shouldn't once it understands me, but it shouldn't once it understands me, but it shouldn't once it understands me, but it's very quick to build that it's very quick to build that it's very quick to build that understanding, and I find I have to give understanding, and I find I have to give understanding, and I find I have to give it less to know it will get things it less to know it will get things it less to know it will get things right. If I give Soul and Fable both the right. If I give Soul and Fable both the right. If I give Soul and Fable both the same two to three sentence prompt, I same two to three sentence prompt, I same two to three sentence prompt, I trust Fable to know what I wanted better trust Fable to know what I wanted better trust Fable to know what I wanted better than Soul did. If I asked for a change than Soul did. If I asked for a change than Soul did. If I asked for a change in the UI that actually requires some in the UI that actually requires some in the UI that actually requires some backend changes, so will get really backend changes, so will get really backend changes, so will get really confused and try to work around the need confused and try to work around the need confused and try to work around the need for the backend change, Babel will just for the backend change, Babel will just for the backend change, Babel will just go do it and let me know in its response go do it and let me know in its response go do it and let me know in its response at the end. Or maybe it'll ask a at the end. Or maybe it'll ask a at the end. Or maybe it'll ask a question to make sure that's what I question to make sure that's what I question to make sure that's what I actually wanted. Obviously, these models actually wanted. Obviously, these models actually wanted. Obviously, these models are just piles of bites that are being are just piles of bites that are being are just piles of bites that are being run by GPUs. They can't really run by GPUs. They can't really run by GPUs. They can't really understand anything. But the feeling of understand anything. But the feeling of understand anything. But the feeling of it understanding is a real difference.

it understanding is a real difference. it understanding is a real difference. And I do feel that difference with And I do feel that difference with And I do feel that difference with Fable. In that same sense, I find that Fable. In that same sense, I find that Fable. In that same sense, I find that it hallucinates way less. It does it hallucinates way less. It does it hallucinates way less. It does occasionally do it. And I've seen plenty occasionally do it. And I've seen plenty occasionally do it. And I've seen plenty of times in reasoning traces where it of times in reasoning traces where it of times in reasoning traces where it got something wrong. And it will say to got something wrong. And it will say to got something wrong. And it will say to itself, I just made that up. I'm going itself, I just made that up. I'm going itself, I just made that up. I'm going to do this other thing instead that was to do this other thing instead that was to do this other thing instead that was actually there, which has been really actually there, which has been really actually there, which has been really nice that it can catch itself. And when nice that it can catch itself. And when nice that it can catch itself. And when you look at certain benchmarks like the you look at certain benchmarks like the you look at certain benchmarks like the omniscience benchmark from artificial omniscience benchmark from artificial omniscience benchmark from artificial analysis, you'll see that Fable has one analysis, you'll see that Fable has one analysis, you'll see that Fable has one of the best scores by far because this of the best scores by far because this of the best scores by far because this test isn't just testing whether the test isn't just testing whether the test isn't just testing whether the model is smart. It's testing how it model is smart. It's testing how it model is smart. It's testing how it behaves when it doesn't have the answer, behaves when it doesn't have the answer, behaves when it doesn't have the answer, when it doesn't know. And when Fable 5 when it doesn't know. And when Fable 5 when it doesn't know. And when Fable 5 doesn't know, it's very quick to admit doesn't know, it's very quick to admit doesn't know, it's very quick to admit it. When Soul doesn't know, it's very it. When Soul doesn't know, it's very it. When Soul doesn't know, it's very quick to work around it or make quick to work around it or make quick to work around it or make something up to unblock itself. It feels something up to unblock itself. It feels something up to unblock itself. It feels like soul was trained so hard on like soul was trained so hard on like soul was trained so hard on unblocking itself that it often ends up unblocking itself that it often ends up unblocking itself that it often ends up behaving in a dumber way as a result. behaving in a dumber way as a result. behaving in a dumber way as a result. This next part's going to hurt me This next part's going to hurt me This next part's going to hurt me because I hate using this word, but but because I hate using this word, but but because I hate using this word, but but we do need to talk about taste a little we do need to talk about taste a little we do need to talk about taste a little bit because fable it has better taste.

bit because fable it has better taste. bit because fable it has better taste. It makes decisions that feel more It makes decisions that feel more It makes decisions that feel more aligned with what I want. And obviously aligned with what I want. And obviously aligned with what I want. And obviously a lot of taste is how well does the a lot of taste is how well does the a lot of taste is how well does the model do the things you like in the way model do the things you like in the way model do the things you like in the way you like and it does them much better. you like and it does them much better. you like and it does them much better. At least for me, this is small things At least for me, this is small things At least for me, this is small things like the copy that it writes. Although like the copy that it writes. Although like the copy that it writes. Although both are pretty cringe at writing, I both are pretty cringe at writing, I both are pretty cringe at writing, I still prefer Kimmy. K2's pros to still prefer Kimmy. K2's pros to still prefer Kimmy. K2's pros to anything these models have. And that's anything these models have. And that's anything these models have. And that's the non-thinking Kimmy. From K21 the non-thinking Kimmy. From K21 the non-thinking Kimmy. From K21 onwards, and especially K25, Kimmy's now onwards, and especially K25, Kimmy's now onwards, and especially K25, Kimmy's now too rled on being good at code and too rled on being good at code and too rled on being good at code and design and not as good at talking. So design and not as good at talking. So design and not as good at talking. So take that as you will. I really liked take that as you will. I really liked take that as you will. I really liked how Kimmy talked with K2. Fable and Soul how Kimmy talked with K2. Fable and Soul how Kimmy talked with K2. Fable and Soul still talk like AI, but Fable does it a still talk like AI, but Fable does it a still talk like AI, but Fable does it a little less cringe. The more important little less cringe. The more important little less cringe. The more important piece with taste though is how it piece with taste though is how it piece with taste though is how it applies the decisions that it makes. applies the decisions that it makes. applies the decisions that it makes. Whether that is in design, whether that Whether that is in design, whether that Whether that is in design, whether that is architecture, whether that is the is architecture, whether that is the is architecture, whether that is the scaffolding for the app you're building scaffolding for the app you're building scaffolding for the app you're building or if it's just the code itself. It or if it's just the code itself. It or if it's just the code itself. It applies its decisions here in more applies its decisions here in more applies its decisions here in more tasteful ways and it makes more tasteful tasteful ways and it makes more tasteful tasteful ways and it makes more tasteful decisions in general. One of the things decisions in general. One of the things decisions in general. One of the things it does much much better is writing less it does much much better is writing less it does much much better is writing less code. I probably should have included code. I probably should have included code. I probably should have included this in the negative section when I was this in the negative section when I was this in the negative section when I was talking about my problems with Soul. My talking about my problems with Soul. My talking about my problems with Soul. My biggest problem with it by far is that biggest problem with it by far is that biggest problem with it by far is that it writes way too much code. It is it writes way too much code. It is it writes way too much code. It is genuinely so frustrating how much code genuinely so frustrating how much code genuinely so frustrating how much code Soul will write in its attempts to solve Soul will write in its attempts to solve Soul will write in its attempts to solve a problem. It's at the point now where a problem. It's at the point now where a problem. It's at the point now where when I'm trying to solve a hard task and when I'm trying to solve a hard task and when I'm trying to solve a hard task and I actually want the result to merge, I I actually want the result to merge, I I actually want the result to merge, I will only use Soul for investigatory will only use Soul for investigatory will only use Soul for investigatory work or as a assigned worker work or as a assigned worker work or as a assigned worker implementing specific things. Fable is implementing specific things. Fable is implementing specific things. Fable is way better at writing the small subset way better at writing the small subset way better at writing the small subset of code that is actually needed to solve

of code that is actually needed to solve of code that is actually needed to solve the problem. This is a thing that I wish the problem. This is a thing that I wish the problem. This is a thing that I wish more benchmarks would measure, but sadly more benchmarks would measure, but sadly more benchmarks would measure, but sadly most benchmarks are measuring how most benchmarks are measuring how most benchmarks are measuring how successful was the model at getting the successful was the model at getting the successful was the model at getting the tests to pass or completing the task as tests to pass or completing the task as tests to pass or completing the task as assigned, not how good is the code and assigned, not how good is the code and assigned, not how good is the code and how likely is it to merge. And while how likely is it to merge. And while how likely is it to merge. And while there's a lot of things I don't like there's a lot of things I don't like there's a lot of things I don't like about the Frontier Code bench from the about the Frontier Code bench from the about the Frontier Code bench from the guys over at Cognition, I do think they guys over at Cognition, I do think they guys over at Cognition, I do think they got this part right. got this part right. got this part right. And a big part of why Opus and Fable And a big part of why Opus and Fable And a big part of why Opus and Fable scored so much better in this bench is scored so much better in this bench is scored so much better in this bench is because it's trying to measure how because it's trying to measure how because it's trying to measure how likely is the code to merge, not just likely is the code to merge, not just likely is the code to merge, not just how successful is the code at passing how successful is the code at passing how successful is the code at passing the test. And as I have experienced many the test. And as I have experienced many the test. And as I have experienced many times now, and I'm sure many of you guys times now, and I'm sure many of you guys times now, and I'm sure many of you guys will see as well, Soul is quick to write will see as well, Soul is quick to write will see as well, Soul is quick to write 10,000 lines of code that aren't really 10,000 lines of code that aren't really 10,000 lines of code that aren't really needed, whereas Fable will write the 100 needed, whereas Fable will write the 100 needed, whereas Fable will write the 100 lines of code that are. And while I have lines of code that are. And while I have lines of code that are. And while I have opened more PRs with soul by quite a opened more PRs with soul by quite a opened more PRs with soul by quite a bit, I have merged way more of the fable bit, I have merged way more of the fable bit, I have merged way more of the fable ones than the soul ones. One more piece, ones than the soul ones. One more piece, ones than the soul ones. One more piece, it's really just one more word that I it's really just one more word that I it's really just one more word that I also feel bad putting in here, but I also feel bad putting in here, but I also feel bad putting in here, but I kind of have to. Fable feels clever in a kind of have to. Fable feels clever in a kind of have to. Fable feels clever in a way that I haven't noticed before. I I way that I haven't noticed before. I I way that I haven't noticed before. I I don't know how to put it, but Fable don't know how to put it, but Fable don't know how to put it, but Fable comes up with solutions that I wouldn't comes up with solutions that I wouldn't comes up with solutions that I wouldn't have come up with in problems that I'm have come up with in problems that I'm have come up with in problems that I'm very experienced in. It has impressed me very experienced in. It has impressed me very experienced in. It has impressed me how often Fable 5 finds a good shortcut how often Fable 5 finds a good shortcut how often Fable 5 finds a good shortcut to a problem. That's why I like it being to a problem. That's why I like it being to a problem. That's why I like it being the model that steers other ones because the model that steers other ones because the model that steers other ones because when it gets back code and it notices when it gets back code and it notices when it gets back code and it notices the code could be simpler, it can the code could be simpler, it can the code could be simpler, it can correct it and tell another sub agent to correct it and tell another sub agent to correct it and tell another sub agent to go fix it. And I love that about Fable.

go fix it. And I love that about Fable. go fix it. And I love that about Fable. It's cleverness. I don't know where it It's cleverness. I don't know where it It's cleverness. I don't know where it comes from. This is a thing they trained comes from. This is a thing they trained comes from. This is a thing they trained it for, if this is just a thing that it for, if this is just a thing that it for, if this is just a thing that comes from it being such a big model, comes from it being such a big model, comes from it being such a big model, but I love when Fable helps me simplify but I love when Fable helps me simplify but I love when Fable helps me simplify problems. And I have used Fable to find problems. And I have used Fable to find problems. And I have used Fable to find and clean up and delete so much code. and clean up and delete so much code. and clean up and delete so much code. It's genuinely really nice. And with It's genuinely really nice. And with It's genuinely really nice. And with that, I will include this because I that, I will include this because I that, I will include this because I guess it does fit. It's a big model. 56 guess it does fit. It's a big model. 56 guess it does fit. It's a big model. 56 is the same size as 55 almost certainly. is the same size as 55 almost certainly. is the same size as 55 almost certainly. It is a different post training. It is It is a different post training. It is It is a different post training. It is unbelievable that OpenAI can take a unbelievable that OpenAI can take a unbelievable that OpenAI can take a model that is as powerful but weird and model that is as powerful but weird and model that is as powerful but weird and bad at staying on task and quick to get bad at staying on task and quick to get bad at staying on task and quick to get its context [ __ ] up like 55 and its context [ __ ] up like 55 and its context [ __ ] up like 55 and through enough abuse of RL can somehow through enough abuse of RL can somehow through enough abuse of RL can somehow make the model perform even close to an make the model perform even close to an make the model perform even close to an allnew bigger pre-training like Fable. allnew bigger pre-training like Fable. allnew bigger pre-training like Fable. It's unbelievable. And the post-training It's unbelievable. And the post-training It's unbelievable. And the post-training team at OpenAI deserves the greatest team at OpenAI deserves the greatest team at OpenAI deserves the greatest praise and reward in the world for what praise and reward in the world for what praise and reward in the world for what they've achieved there. It is truly they've achieved there. It is truly they've achieved there. It is truly unbelievable. They could take an unbelievable. They could take an unbelievable. They could take an existing smaller model and RL it into existing smaller model and RL it into existing smaller model and RL it into one of the best models ever made. The one of the best models ever made. The one of the best models ever made. The second best model ever made to be second best model ever made to be second best model ever made to be specific. It is insane what they did specific. It is insane what they did specific. It is insane what they did here. But Fable 5 is a new architecture.

here. But Fable 5 is a new architecture. here. But Fable 5 is a new architecture. It's a new model that is bigger and has It's a new model that is bigger and has It's a new model that is bigger and has more information. It is capable of more more information. It is capable of more more information. It is capable of more as a result. That all said, bigger model as a result. That all said, bigger model as a result. That all said, bigger model doesn't come cost free. In fact, bigger doesn't come cost free. In fact, bigger doesn't come cost free. In fact, bigger model costs way more because the model model costs way more because the model model costs way more because the model is bigger. It's more expensive to run. is bigger. It's more expensive to run. is bigger. It's more expensive to run. And because they haven't rldled it to be And because they haven't rldled it to be And because they haven't rldled it to be way more efficient in reasoning like way more efficient in reasoning like way more efficient in reasoning like OpenAI has done, they have made OpenAI has done, they have made OpenAI has done, they have made improvements here, but it's still not improvements here, but it's still not improvements here, but it's still not the same level, it ends up burning more the same level, it ends up burning more the same level, it ends up burning more tokens at a higher cost in generating tokens at a higher cost in generating tokens at a higher cost in generating them slower, resulting in a way slower them slower, resulting in a way slower them slower, resulting in a way slower endto-end experience, which is one of endto-end experience, which is one of endto-end experience, which is one of the things I like Soul for so much. It the things I like Soul for so much. It the things I like Soul for so much. It also has a habit of giving up slightly also has a habit of giving up slightly also has a habit of giving up slightly too early sometimes where it should take too early sometimes where it should take too early sometimes where it should take the next step, but it doesn't because the next step, but it doesn't because the next step, but it doesn't because it's often just trying to be clever and it's often just trying to be clever and it's often just trying to be clever and find a work. It's like, what if I just find a work. It's like, what if I just find a work. It's like, what if I just don't do that thing? And I'm like, wait, don't do that thing? And I'm like, wait, don't do that thing? And I'm like, wait, you probably should go do that thing, you probably should go do that thing, you probably should go do that thing, though. This is part of its cleverness though. This is part of its cleverness though. This is part of its cleverness and its intelligence. And also that it and its intelligence. And also that it and its intelligence. And also that it doesn't have this absurdly overeagerness doesn't have this absurdly overeagerness doesn't have this absurdly overeagerness built in. But it is a little quicker to built in. But it is a little quicker to built in. But it is a little quicker to give up on a task than Soul is. And it give up on a task than Soul is. And it give up on a task than Soul is. And it also can sometimes fix it on the wrong also can sometimes fix it on the wrong also can sometimes fix it on the wrong thing and fall down rabbit holes. Not thing and fall down rabbit holes. Not thing and fall down rabbit holes. Not that Soul can't. They both do this to an that Soul can't. They both do this to an that Soul can't. They both do this to an extent. I've had both make really bad extent. I've had both make really bad extent. I've had both make really bad suggestions many times. I have a video suggestions many times. I have a video suggestions many times. I have a video coming out soon about a bug that Fable coming out soon about a bug that Fable coming out soon about a bug that Fable and Soul caused that neither could solve and Soul caused that neither could solve and Soul caused that neither could solve because they both suck at performance because they both suck at performance because they both suck at performance analysis in Chrome, especially when it's analysis in Chrome, especially when it's analysis in Chrome, especially when it's GPU related. But yeah, Fable will give GPU related. But yeah, Fable will give GPU related. But yeah, Fable will give up early a little bit too often. And up early a little bit too often. And up early a little bit too often. And sometimes it's not actually Fable's sometimes it's not actually Fable's sometimes it's not actually Fable's fault because sometimes it gets fault because sometimes it gets fault because sometimes it gets rerouted. And the rerouting, it just rerouted. And the rerouting, it just rerouted. And the rerouting, it just sucks. I had some problems with the sucks. I had some problems with the sucks. I had some problems with the rerouting on 56 when it came out, but it

rerouting on 56 when it came out, but it rerouting on 56 when it came out, but it seems to have been toned down quite a seems to have been toned down quite a seems to have been toned down quite a bit and I haven't experienced it at all bit and I haven't experienced it at all bit and I haven't experienced it at all since day one. Whereas with Fable 5, it since day one. Whereas with Fable 5, it since day one. Whereas with Fable 5, it does happen a bit. It's not a lot. I does happen a bit. It's not a lot. I does happen a bit. It's not a lot. I haven't seen it for a few days, but haven't seen it for a few days, but haven't seen it for a few days, but Fable is still a little quick to reroute Fable is still a little quick to reroute Fable is still a little quick to reroute you to Opus for anything vaguely you to Opus for anything vaguely you to Opus for anything vaguely cryptography related, and I've seen cryptography related, and I've seen cryptography related, and I've seen people get rerouted multiple times in people get rerouted multiple times in people get rerouted multiple times in one request. Toucan here is an one request. Toucan here is an one request. Toucan here is an exanthropic employee, and he said the exanthropic employee, and he said the exanthropic employee, and he said the following. He had a Fable 5 chat fall following. He had a Fable 5 chat fall following. He had a Fable 5 chat fall back to Opus 48, which fell back to back to Opus 48, which fell back to back to Opus 48, which fell back to Sonnet 46, which then fell back to Haiku Sonnet 46, which then fell back to Haiku Sonnet 46, which then fell back to Haiku 45. That sucks. And this is a uniquely 45. That sucks. And this is a uniquely 45. That sucks. And this is a uniquely anthropic problem. These are while soul anthropic problem. These are while soul anthropic problem. These are while soul will now reroute which sucks that there will now reroute which sucks that there will now reroute which sucks that there is no Frontier model that won't is no Frontier model that won't is no Frontier model that won't occasionally reroute you to something occasionally reroute you to something occasionally reroute you to something dumber. It happens a lot less and Fable dumber. It happens a lot less and Fable dumber. It happens a lot less and Fable is a little too aggressive with this to is a little too aggressive with this to is a little too aggressive with this to the point where it is uh frustrating to the point where it is uh frustrating to the point where it is uh frustrating to use at points. And one more big negative use at points. And one more big negative use at points. And one more big negative I don't want to put this other than I don't want to put this other than I don't want to put this other than exclusivity. The codec subscriptions are exclusivity. The codec subscriptions are exclusivity. The codec subscriptions are relatively generous. They kind of just relatively generous. They kind of just relatively generous. They kind of just let you use codecs in whatever tools you let you use codecs in whatever tools you let you use codecs in whatever tools you prefer. The cloud subs are much less prefer. The cloud subs are much less prefer. The cloud subs are much less generous. basically forcing you into generous. basically forcing you into generous. basically forcing you into cloud code and their harnesses in other cloud code and their harnesses in other cloud code and their harnesses in other ways. They did officially support using ways. They did officially support using ways. They did officially support using your sub for local use cases using their your sub for local use cases using their your sub for local use cases using their agents SDK which is a closed source agents SDK which is a closed source agents SDK which is a closed source TypeScript SDK that provides cloud code TypeScript SDK that provides cloud code TypeScript SDK that provides cloud code with programmatic access. They had with programmatic access. They had with programmatic access. They had announced they were going to ban people announced they were going to ban people announced they were going to ban people from that but they seem to have walked from that but they seem to have walked from that but they seem to have walked that back announcing that they canled that back announcing that they canled that back announcing that they canled that plan change and they haven't put up that plan change and they haven't put up that plan change and they haven't put up their new plan just yet. So thankfully their new plan just yet. So thankfully their new plan just yet. So thankfully tools like T3 Code are still supported.

tools like T3 Code are still supported. tools like T3 Code are still supported. Other tools that fully change out the Other tools that fully change out the Other tools that fully change out the harness like pi or open code aren't yet. harness like pi or open code aren't yet. harness like pi or open code aren't yet. But since T3 code is calling those other But since T3 code is calling those other But since T3 code is calling those other harnesses because T3 code isn't a harnesses because T3 code isn't a harnesses because T3 code isn't a harness. It doesn't call the model harness. It doesn't call the model harness. It doesn't call the model directly. It calls the harness whether directly. It calls the harness whether directly. It calls the harness whether it's codec quad code open code or it's codec quad code open code or it's codec quad code open code or whatever else and then the model gets whatever else and then the model gets whatever else and then the model gets called through that through their called through that through their called through that through their official code. We are still good. But if official code. We are still good. But if official code. We are still good. But if you want to use your Fable subscription you want to use your Fable subscription you want to use your Fable subscription for other things, you're kind of for other things, you're kind of for other things, you're kind of screwed. I use my subscription with 5.6 screwed. I use my subscription with 5.6 screwed. I use my subscription with 5.6 for a lot of different things. I use it for a lot of different things. I use it for a lot of different things. I use it for some automation work to keep track for some automation work to keep track for some automation work to keep track of all of my PRs. I use it inside of of all of my PRs. I use it inside of of all of my PRs. I use it inside of other harnesses like PI and Open Code other harnesses like PI and Open Code other harnesses like PI and Open Code all of the time. I use it inside of my all of the time. I use it inside of my all of the time. I use it inside of my Hermes agent and my Open Claw for just Hermes agent and my Open Claw for just Hermes agent and my Open Claw for just doing all sorts of tasks for my life. I doing all sorts of tasks for my life. I doing all sorts of tasks for my life. I know Ben, my podcast host and channel know Ben, my podcast host and channel know Ben, my podcast host and channel manager, uses his subscription to power manager, uses his subscription to power manager, uses his subscription to power a bunch of bots that we have in Discord a bunch of bots that we have in Discord a bunch of bots that we have in Discord for managing all of our channels and for managing all of our channels and for managing all of our channels and media and sponsors and whatnot. All of media and sponsors and whatnot. All of media and sponsors and whatnot. All of those things are things we can do on the those things are things we can do on the those things are things we can do on the single $200 plan with Soul. And our single $200 plan with Soul. And our single $200 plan with Soul. And our insane inference just barely costs insane inference just barely costs insane inference just barely costs anything. We pay that one fee and get anything. We pay that one fee and get anything. We pay that one fee and get practically unlimited usage. Obviously, practically unlimited usage. Obviously, practically unlimited usage. Obviously, if you're using Facet Ultra, that if you're using Facet Ultra, that if you're using Facet Ultra, that doesn't feel unlimited anymore. Go check doesn't feel unlimited anymore. Go check doesn't feel unlimited anymore. Go check out my other video about all that. But out my other video about all that. But out my other video about all that. But for real day-to-day use, Solo has been for real day-to-day use, Solo has been for real day-to-day use, Solo has been great. And I really like it for those great. And I really like it for those great. And I really like it for those use cases like a chatbot or something use cases like a chatbot or something use cases like a chatbot or something inside of my Hermes agent because it's inside of my Hermes agent because it's inside of my Hermes agent because it's so much faster that it gets a good so much faster that it gets a good so much faster that it gets a good response quicker. In Fable, you're just response quicker. In Fable, you're just response quicker. In Fable, you're just kind of sitting and waiting for. But you kind of sitting and waiting for. But you kind of sitting and waiting for. But you also have to pay full API price for also have to pay full API price for also have to pay full API price for Fable if you use it at those places Fable if you use it at those places Fable if you use it at those places because you can't get the subsidization because you can't get the subsidization because you can't get the subsidization that you get through the subscription that you get through the subscription that you get through the subscription outside of a quad code directly.

outside of a quad code directly. outside of a quad code directly. Thankfully, T4 code is still supported. Thankfully, T4 code is still supported. Thankfully, T4 code is still supported. If you want a better UI, you can use If you want a better UI, you can use If you want a better UI, you can use that. But generally speaking, Fable 5 is that. But generally speaking, Fable 5 is that. But generally speaking, Fable 5 is very exclusive to your sub. At least it very exclusive to your sub. At least it very exclusive to your sub. At least it is for one or two more days because it's is for one or two more days because it's is for one or two more days because it's about to be removed. They've delayed the about to be removed. They've delayed the about to be removed. They've delayed the removal three separate times, but it is removal three separate times, but it is removal three separate times, but it is still about to get removed. We are going still about to get removed. We are going still about to get removed. We are going to lose access to Fable inside of our to lose access to Fable inside of our to lose access to Fable inside of our subscription plans very soon. And even subscription plans very soon. And even subscription plans very soon. And even now where we do have it, Fable only can now where we do have it, Fable only can now where we do have it, Fable only can be used for half of your subscription be used for half of your subscription be used for half of your subscription limit. Your Fable limit will get hit 2x limit. Your Fable limit will get hit 2x limit. Your Fable limit will get hit 2x faster than everything else because it faster than everything else because it faster than everything else because it is just so much compute that it's is just so much compute that it's is just so much compute that it's killing them. My guess is that their killing them. My guess is that their killing them. My guess is that their plan is to put out Opus 5, which should plan is to put out Opus 5, which should plan is to put out Opus 5, which should be a smaller model with similar be a smaller model with similar be a smaller model with similar capability, and then convince people on capability, and then convince people on capability, and then convince people on the subscription plan to go use that the subscription plan to go use that the subscription plan to go use that instead. But it's not going to have a instead. But it's not going to have a instead. But it's not going to have a lot of the benefits I like. It's not lot of the benefits I like. It's not lot of the benefits I like. It's not going to be that big model that has the going to be that big model that has the going to be that big model that has the same taste and is carefully applying same taste and is carefully applying same taste and is carefully applying each line of code instead of writing a each line of code instead of writing a each line of code instead of writing a ton of slop. It might be really good at ton of slop. It might be really good at ton of slop. It might be really good at orchestrating sub agents still. It orchestrating sub agents still. It orchestrating sub agents still. It probably will be, but I'm really not probably will be, but I'm really not probably will be, but I'm really not thinking it's going to be as capable as thinking it's going to be as capable as thinking it's going to be as capable as Fable. So when you combine the Fable. So when you combine the Fable. So when you combine the government ban of Fable with the government ban of Fable with the government ban of Fable with the absurdity of how it is coming into and absurdity of how it is coming into and absurdity of how it is coming into and out of subscriptions with the fact that out of subscriptions with the fact that out of subscriptions with the fact that you can even use the subscription you can even use the subscription you can even use the subscription outside of their services and then you outside of their services and then you outside of their services and then you look at the cost of what it looks like look at the cost of what it looks like look at the cost of what it looks like to actually run Fable in real world to actually run Fable in real world to actually run Fable in real world work, it starts to hurt. That said, when work, it starts to hurt. That said, when work, it starts to hurt. That said, when I want my changes to land, Fable is the I want my changes to land, Fable is the I want my changes to land, Fable is the first model I trust to get the code so first model I trust to get the code so first model I trust to get the code so good first try that I don't feel bad good first try that I don't feel bad good first try that I don't feel bad just immediately filing the PR. Every just immediately filing the PR. Every just immediately filing the PR. Every other model generates meaningfully more

other model generates meaningfully more other model generates meaningfully more slop code that needs a lot more slop code that needs a lot more slop code that needs a lot more handholding to get into a good state. handholding to get into a good state. handholding to get into a good state. Fable carefully, concisely applies the Fable carefully, concisely applies the Fable carefully, concisely applies the right diff so often that I reach for it right diff so often that I reach for it right diff so often that I reach for it every time I actually want to land my every time I actually want to land my every time I actually want to land my changes. And while Soul is an incredibly changes. And while Soul is an incredibly changes. And while Soul is an incredibly capable, powerful, efficient, just capable, powerful, efficient, just capable, powerful, efficient, just awesome to use model, it is kind of awesome to use model, it is kind of awesome to use model, it is kind of crazy that when I want to land code, crazy that when I want to land code, crazy that when I want to land code, there's only one of these two that I there's only one of these two that I there's only one of these two that I immediately reach for, and it is almost immediately reach for, and it is almost immediately reach for, and it is almost always fable. I do have a way to compare always fable. I do have a way to compare always fable. I do have a way to compare these that I think makes these these that I think makes these these that I think makes these behavioral differences make much more behavioral differences make much more behavioral differences make much more sense, but I'm not the one who came up sense, but I'm not the one who came up sense, but I'm not the one who came up with it, so I want to find the original with it, so I want to find the original with it, so I want to find the original source and share it with y'all. It is source and share it with y'all. It is source and share it with y'all. It is this post by Peter Gustaf. I think it this post by Peter Gustaf. I think it this post by Peter Gustaf. I think it does a phenomenal job of explaining the does a phenomenal job of explaining the does a phenomenal job of explaining the difference. First, he says they're not difference. First, he says they're not difference. First, he says they're not easy models to compare, and I couldn't easy models to compare, and I couldn't easy models to compare, and I couldn't agree more. They are similarly capable agree more. They are similarly capable agree more. They are similarly capable and similarly intelligent, but the and similarly intelligent, but the and similarly intelligent, but the feeling of them is so different. It's so feeling of them is so different. It's so feeling of them is so different. It's so different. His overall feel is that different. His overall feel is that different. His overall feel is that Fable is a wise owl who is very Fable is a wise owl who is very Fable is a wise owl who is very thoughtful and very well spoken. 56 Soul thoughtful and very well spoken. 56 Soul thoughtful and very well spoken. 56 Soul is like a Rottweiler who will grab the is like a Rottweiler who will grab the is like a Rottweiler who will grab the problem by the throat and not let go problem by the throat and not let go problem by the throat and not let go until it's done. Spot [ __ ] on. This until it's done. Spot [ __ ] on. This until it's done. Spot [ __ ] on. This is exactly how I feel about using these is exactly how I feel about using these is exactly how I feel about using these two models. Fable is a fundamentally two models. Fable is a fundamentally two models. Fable is a fundamentally smarter model even at low reasoning can smarter model even at low reasoning can smarter model even at low reasoning can be very insightful and it writes in a be very insightful and it writes in a be very insightful and it writes in a clear and compelling way. 56 soul on the clear and compelling way. 56 soul on the clear and compelling way. 56 soul on the other hand is extremely diligent. Can other hand is extremely diligent. Can other hand is extremely diligent. Can give a list of eight things to do and give a list of eight things to do and give a list of eight things to do and you will be sure that all of them will you will be sure that all of them will you will be sure that all of them will be done. Babel feels more arrogant to be done. Babel feels more arrogant to be done. Babel feels more arrogant to him. He was trying to get it to build a

him. He was trying to get it to build a him. He was trying to get it to build a new benchmark. 56 worked between 6 hours new benchmark. 56 worked between 6 hours new benchmark. 56 worked between 6 hours and 2 days. He tried several times to and 2 days. He tried several times to and 2 days. He tried several times to build it and it came back with a very build it and it came back with a very build it and it came back with a very thoroughly tested working bench. Fable thoroughly tested working bench. Fable thoroughly tested working bench. Fable came back within 40 minutes twice and came back within 40 minutes twice and came back within 40 minutes twice and the benchmark sounded smart, but it was the benchmark sounded smart, but it was the benchmark sounded smart, but it was ultimately a vibe based slop or bench. ultimately a vibe based slop or bench. ultimately a vibe based slop or bench. And since it was Fable's vibes that was And since it was Fable's vibes that was And since it was Fable's vibes that was doing the judging, it decided that it doing the judging, it decided that it doing the judging, it decided that it was good to go. Of course, it was giving was good to go. Of course, it was giving was good to go. Of course, it was giving Fable 100% scores constantly. This is Fable 100% scores constantly. This is Fable 100% scores constantly. This is actually an interesting thing that actually an interesting thing that actually an interesting thing that happened to me during testing. When Opus happened to me during testing. When Opus happened to me during testing. When Opus 48 came out, I had just started testing 48 came out, I had just started testing 48 came out, I had just started testing 56 and I had 48 review some of the code 56 and I had 48 review some of the code 56 and I had 48 review some of the code that I wrote with 56 and it glazed the that I wrote with 56 and it glazed the that I wrote with 56 and it glazed the [ __ ] out of it. Opus48 was in awe of the [ __ ] out of it. Opus48 was in awe of the [ __ ] out of it. Opus48 was in awe of the work that 56 was doing. Then I got Fable work that 56 was doing. Then I got Fable work that 56 was doing. Then I got Fable about a week later and I had Fable about a week later and I had Fable about a week later and I had Fable review the same code and I was like, review the same code and I was like, review the same code and I was like, "Yeah, this is fine, but that's sloppy. "Yeah, this is fine, but that's sloppy. "Yeah, this is fine, but that's sloppy. That's a mess. That's not necessary. I That's a mess. That's not necessary. I That's a mess. That's not necessary. I have no idea why that is there. We can have no idea why that is there. We can have no idea why that is there. We can clean this up." It was insane to see clean this up." It was insane to see clean this up." It was insane to see Fable much more capable of digging into Fable much more capable of digging into Fable much more capable of digging into the problems with Soul's code than Opus the problems with Soul's code than Opus the problems with Soul's code than Opus was. That said, when I had 56 review its was. That said, when I had 56 review its was. That said, when I had 56 review its own code in different threads, of own code in different threads, of own code in different threads, of course, it was very critical of it. Soul course, it was very critical of it. Soul course, it was very critical of it. Soul can be critical of Soul's work. It's can be critical of Soul's work. It's can be critical of Soul's work. It's surprisingly good at that. I could never surprisingly good at that. I could never surprisingly good at that. I could never get Fable to say anything negative about get Fable to say anything negative about get Fable to say anything negative about work that was done by Fable. Even with work that was done by Fable. Even with work that was done by Fable. Even with new threads, new contexts, new new threads, new contexts, new new threads, new contexts, new everything, Fable almost feels like it everything, Fable almost feels like it everything, Fable almost feels like it can smell work it did and gives it the can smell work it did and gives it the can smell work it did and gives it the the craziest glaze fest it possibly can.

the craziest glaze fest it possibly can. the craziest glaze fest it possibly can. I would never trust Fable to make a I would never trust Fable to make a I would never trust Fable to make a benchmark for these reasons cuz it is benchmark for these reasons cuz it is benchmark for these reasons cuz it is going to promote itself to the [ __ ] going to promote itself to the [ __ ] going to promote itself to the [ __ ] moon. Back to this comparison, I think moon. Back to this comparison, I think moon. Back to this comparison, I think this is really good. Fable still crafts this is really good. Fable still crafts this is really good. Fable still crafts better UIs from scratch. The flow of the better UIs from scratch. The flow of the better UIs from scratch. The flow of the app would probably be nicer if you start app would probably be nicer if you start app would probably be nicer if you start with it, but Fable will miss key things with it, but Fable will miss key things with it, but Fable will miss key things while Soul doesn't. I actually just while Soul doesn't. I actually just while Soul doesn't. I actually just experienced this a few times when I was experienced this a few times when I was experienced this a few times when I was working earlier today. Actually, when I working earlier today. Actually, when I working earlier today. Actually, when I was doing this sidebar overhaul, I kept was doing this sidebar overhaul, I kept was doing this sidebar overhaul, I kept insisting to Fable that it's important insisting to Fable that it's important insisting to Fable that it's important we know which project is open both in we know which project is open both in we know which project is open both in the sidebar and in the thread. And then the sidebar and in the thread. And then the sidebar and in the thread. And then I would do one follow-up to make a I would do one follow-up to make a I would do one follow-up to make a change and it would make changes that change and it would make changes that change and it would make changes that removed the name of the project from removed the name of the project from removed the name of the project from both the sidebar and the thread. And both the sidebar and the thread. And both the sidebar and the thread. And then I would have to insist, no, make then I would have to insist, no, make then I would have to insist, no, make sure this is included. And then it would sure this is included. And then it would sure this is included. And then it would go apply it really poorly and say, make go apply it really poorly and say, make go apply it really poorly and say, make it more tasteful. then another change it more tasteful. then another change it more tasteful. then another change and it would disappear again. So, I and it would disappear again. So, I and it would disappear again. So, I ended up having to like write a really ended up having to like write a really ended up having to like write a really strict message. Let me see if I can find strict message. Let me see if I can find strict message. Let me see if I can find it in my history. Actually, here I said, it in my history. Actually, here I said, it in my history. Actually, here I said, I'm concerned we may have made it too I'm concerned we may have made it too I'm concerned we may have made it too difficult in general to see what project difficult in general to see what project difficult in general to see what project you're actually in, which sucks because you're actually in, which sucks because you're actually in, which sucks because that's like priority zero. And I [ __ ] that's like priority zero. And I [ __ ] that's like priority zero. And I [ __ ] you not, it screwed up again. It tried you not, it screwed up again. It tried you not, it screwed up again. It tried using color boxes, which were hideous.

using color boxes, which were hideous. using color boxes, which were hideous. So, I told it try again. I don't like So, I told it try again. I don't like So, I told it try again. I don't like the grouping cuz this point it like the grouping cuz this point it like the grouping cuz this point it like tried grouping based on project which we tried grouping based on project which we tried grouping based on project which we had moved away from. that I really want had moved away from. that I really want had moved away from. that I really want the threads to be fluid and statically the threads to be fluid and statically the threads to be fluid and statically ordered from when they're open so ordered from when they're open so ordered from when they're open so there's less moving around on the there's less moving around on the there's less moving around on the screen. And then it again removed all screen. And then it again removed all screen. And then it again removed all indication of what project was open. indication of what project was open. indication of what project was open. There was no way to know and I was like There was no way to know and I was like There was no way to know and I was like so pissed off that I wrote a little so pissed off that I wrote a little so pissed off that I wrote a little essay. Let's get a few things clear essay. Let's get a few things clear essay. Let's get a few things clear because you keep messing them up. I will because you keep messing them up. I will because you keep messing them up. I will not compromise on the following. One, not compromise on the following. One, not compromise on the following. One, indication of which project in sidebar indication of which project in sidebar indication of which project in sidebar threads. Two, indication of which threads. Two, indication of which threads. Two, indication of which project in the currently open thread project in the currently open thread project in the currently open thread thread. Three, using favicons where thread. Three, using favicons where thread. Three, using favicons where available for threads. And four, no available for threads. And four, no available for threads. And four, no grouping by project and sidebar. And I grouping by project and sidebar. And I grouping by project and sidebar. And I was so annoyed that I didn't trust it to was so annoyed that I didn't trust it to was so annoyed that I didn't trust it to do one pass. So I did my favorite trick. do one pass. So I did my favorite trick. do one pass. So I did my favorite trick. I asked it to go make five mocks I asked it to go make five mocks I asked it to go make five mocks somewhere else that I could go through somewhere else that I could go through somewhere else that I could go through in order to get this right. And once I in order to get this right. And once I in order to get this right. And once I saw all of these mocks, I found some saw all of these mocks, I found some saw all of these mocks, I found some parts that were good. In fact, I think parts that were good. In fact, I think parts that were good. In fact, I think it was only one that was good. Is this it was only one that was good. Is this it was only one that was good. Is this two is great. Do that is all I said. And two is great. Do that is all I said. And two is great. Do that is all I said. And then it went and did it. And it came out then it went and did it. And it came out then it went and did it. And it came out really good. But like that's what I had really good. But like that's what I had really good. But like that's what I had to do. I had to brutalize this model to do. I had to brutalize this model to do. I had to brutalize this model into actually listening to me because it into actually listening to me because it into actually listening to me because it kept forgetting the details that kept forgetting the details that kept forgetting the details that mattered because it's trying too hard to mattered because it's trying too hard to mattered because it's trying too hard to assume intent and not to listen. This is assume intent and not to listen. This is assume intent and not to listen. This is one of the biggest differences I have one of the biggest differences I have one of the biggest differences I have felt between the two. Soul doesn't read felt between the two. Soul doesn't read felt between the two. Soul doesn't read between the lines. It reads what you between the lines. It reads what you between the lines. It reads what you said and it does exactly that, even if said and it does exactly that, even if said and it does exactly that, even if that means destroying things in the that means destroying things in the that means destroying things in the process. Babel will read between the process. Babel will read between the process. Babel will read between the lines a lot, which is often good because lines a lot, which is often good because lines a lot, which is often good because it will come up with better solutions

it will come up with better solutions it will come up with better solutions because it's trying to get to the intent because it's trying to get to the intent because it's trying to get to the intent of what you said, not the specific of what you said, not the specific of what you said, not the specific details of what you said. But that also details of what you said. But that also details of what you said. But that also means it could sometimes miss the means it could sometimes miss the means it could sometimes miss the details you're saying in its pursuit of details you're saying in its pursuit of details you're saying in its pursuit of intent. It loses track of what you intent. It loses track of what you intent. It loses track of what you actually care about. But again, you can actually care about. But again, you can actually care about. But again, you can brutalize it into behaving by using brutalize it into behaving by using brutalize it into behaving by using capital letters and lists saying, "No, capital letters and lists saying, "No, capital letters and lists saying, "No, you have to do these things. No you have to do these things. No you have to do these things. No compromise." But like I had to do that. compromise." But like I had to do that. compromise." But like I had to do that. So, as Peter said here, Fable is way So, as Peter said here, Fable is way So, as Peter said here, Fable is way better at creating a UI from scratch and better at creating a UI from scratch and better at creating a UI from scratch and making things that look good, but 56 making things that look good, but 56 making things that look good, but 56 will actually complete the task when will actually complete the task when will actually complete the task when given the details. For writing, Fable's given the details. For writing, Fable's given the details. For writing, Fable's better hands down. Soul feels quite better hands down. Soul feels quite better hands down. Soul feels quite different to align to what he wants to different to align to what he wants to different to align to what he wants to say or explain things to him simply, say or explain things to him simply, say or explain things to him simply, though he does think pro writes clear. I though he does think pro writes clear. I though he does think pro writes clear. I agree. You still can't use pro in code agree. You still can't use pro in code agree. You still can't use pro in code sadly, but it is a clear writer for sadly, but it is a clear writer for sadly, but it is a clear writer for sure. Robustness and reliability. This sure. Robustness and reliability. This sure. Robustness and reliability. This is where I think Soul wins for me hands is where I think Soul wins for me hands is where I think Soul wins for me hands down. Fable seems to do things of high down. Fable seems to do things of high down. Fable seems to do things of high quality, but he can never relax with it. quality, but he can never relax with it. quality, but he can never relax with it. almost always misses something with five almost always misses something with five almost always misses something with five six this almost never happens. I agree. six this almost never happens. I agree. six this almost never happens. I agree. But like you have to understand with any But like you have to understand with any But like you have to understand with any given task there is the perfect 100% given task there is the perfect 100% given task there is the perfect 100% version where everything is solved and version where everything is solved and version where everything is solved and nothing extra is done. There's the 80% nothing extra is done. There's the 80% nothing extra is done. There's the 80% version where it is the simplified thing version where it is the simplified thing version where it is the simplified thing that gets most of the problem solved but that gets most of the problem solved but that gets most of the problem solved but misses a few edges. And then there is misses a few edges. And then there is misses a few edges. And then there is the overdone extra version where you do the overdone extra version where you do the overdone extra version where you do 300% more code, a shitload of smoke 300% more code, a shitload of smoke 300% more code, a shitload of smoke tests, and all these other things that tests, and all these other things that tests, and all these other things that don't really matter. Babel skews between don't really matter. Babel skews between don't really matter. Babel skews between like 80 and 110% from my experience like 80 and 110% from my experience like 80 and 110% from my experience where it just barely misses a few things where it just barely misses a few things where it just barely misses a few things or does a little too much extra. Soul's or does a little too much extra. Soul's or does a little too much extra. Soul's baseline is like 150%. It always does a

baseline is like 150%. It always does a baseline is like 150%. It always does a little too much and it sometimes does little too much and it sometimes does little too much and it sometimes does way too much. I've been playing with way too much. I've been playing with way too much. I've been playing with skills and plugins and things to try and skills and plugins and things to try and skills and plugins and things to try and get Soul to not do that. Even just get Soul to not do that. Even just get Soul to not do that. Even just asking it at the end of every change, asking it at the end of every change, asking it at the end of every change, how can we simplify this? How can we how can we simplify this? How can we how can we simplify this? How can we make this PR as minimal and elegant as make this PR as minimal and elegant as make this PR as minimal and elegant as possible while still achieving our possible while still achieving our possible while still achieving our goals? and it will usually find a few goals? and it will usually find a few goals? and it will usually find a few things, but honestly, what I do nowadays things, but honestly, what I do nowadays things, but honestly, what I do nowadays is I have Fable go in and clean up is I have Fable go in and clean up is I have Fable go in and clean up instead. Peter has a list of other instead. Peter has a list of other instead. Peter has a list of other things he prefers with Soul. Things like things he prefers with Soul. Things like things he prefers with Soul. Things like video editing, which it actually can video editing, which it actually can video editing, which it actually can kind of do now. I don't really agree. kind of do now. I don't really agree. kind of do now. I don't really agree. I've seen it do some surprisingly good I've seen it do some surprisingly good I've seen it do some surprisingly good stuff, like creating the XML files that stuff, like creating the XML files that stuff, like creating the XML files that we use in Final Cut. I know that Jeff, we use in Final Cut. I know that Jeff, we use in Final Cut. I know that Jeff, my editor phase, has had a lot of fun my editor phase, has had a lot of fun my editor phase, has had a lot of fun with that lately, but it sometimes with that lately, but it sometimes with that lately, but it sometimes creates corrupted files, too, and then creates corrupted files, too, and then creates corrupted files, too, and then you have to go do it from scratch you have to go do it from scratch you have to go do it from scratch anyways. It also has no taste with video anyways. It also has no taste with video anyways. It also has no taste with video editing at all. It can just mark things editing at all. It can just mark things editing at all. It can just mark things which does make the editing process by which does make the editing process by which does make the editing process by hand easier. I'm interested to actually hand easier. I'm interested to actually hand easier. I'm interested to actually edit video though. The computer use edit video though. The computer use edit video though. The computer use however is unbelievable. I still cannot however is unbelievable. I still cannot however is unbelievable. I still cannot fathom how crazy the level up is here fathom how crazy the level up is here fathom how crazy the level up is here and how capable 56 is. I often will and how capable 56 is. I often will and how capable 56 is. I often will leave my personal computer with codeex leave my personal computer with codeex leave my personal computer with codeex open. Sorry, with chat GPT in codeex open. Sorry, with chat GPT in codeex open. Sorry, with chat GPT in codeex mode open just so that I can get things mode open just so that I can get things mode open just so that I can get things done on my computer when I'm on my phone done on my computer when I'm on my phone done on my computer when I'm on my phone remotely. It's so nice to have that deep remotely. It's so nice to have that deep remotely. It's so nice to have that deep computer use capability where I can like computer use capability where I can like computer use capability where I can like go into my browser and do things, find a go into my browser and do things, find a go into my browser and do things, find a file in my folders and drag it over file in my folders and drag it over file in my folders and drag it over somewhere else. Like it can complete somewhere else. Like it can complete somewhere else. Like it can complete work on my machine without me there.

work on my machine without me there. work on my machine without me there. It's been really nice. It's also really It's been really nice. It's also really It's been really nice. It's also really good at sub agents. I will say Fable's good at sub agents. I will say Fable's good at sub agents. I will say Fable's also quite good at them too, especially also quite good at them too, especially also quite good at them too, especially workflows. It's good at adhering to workflows. It's good at adhering to workflows. It's good at adhering to existing code patterns, which is really existing code patterns, which is really existing code patterns, which is really nice as well. I would not trust 56 to nice as well. I would not trust 56 to nice as well. I would not trust 56 to create a new project from scratch, but create a new project from scratch, but create a new project from scratch, but it will follow the patterns in the it will follow the patterns in the it will follow the patterns in the codebase better overall too much so codebase better overall too much so codebase better overall too much so sometimes where it'll make things bigger sometimes where it'll make things bigger sometimes where it'll make things bigger than it needs to, but it does really than it needs to, but it does really than it needs to, but it does really just fit into your codebase well. It just fit into your codebase well. It just fit into your codebase well. It doesn't get mad at you when the patterns doesn't get mad at you when the patterns doesn't get mad at you when the patterns aren't what it likes. Fable will talk a aren't what it likes. Fable will talk a aren't what it likes. Fable will talk a lot of [ __ ] about your code. 56 will lot of [ __ ] about your code. 56 will lot of [ __ ] about your code. 56 will just perfectly match the patterns as just perfectly match the patterns as just perfectly match the patterns as they stand. It still leans towards they stand. It still leans towards they stand. It still leans towards writing TypeScript like it's a Python writing TypeScript like it's a Python writing TypeScript like it's a Python dev. I've beat that out of it with a dev. I've beat that out of it with a dev. I've beat that out of it with a stick, but it loves oneline functions stick, but it loves oneline functions stick, but it loves oneline functions that are just typ casting and [ __ ] Not that are just typ casting and [ __ ] Not that are just typ casting and [ __ ] Not my favorite, but you get the idea. It's my favorite, but you get the idea. It's my favorite, but you get the idea. It's getting better at research overall. It getting better at research overall. It getting better at research overall. It has some bad patterns, like it's too has some bad patterns, like it's too has some bad patterns, like it's too tactical, but it's more steerable to be tactical, but it's more steerable to be tactical, but it's more steerable to be a good researcher. This I believe, a good researcher. This I believe, a good researcher. This I believe, especially with pro, but I have found especially with pro, but I have found especially with pro, but I have found that the insights I get from data or that the insights I get from data or that the insights I get from data or just like giving the model access to my just like giving the model access to my just like giving the model access to my post hog, I get better insights out of post hog, I get better insights out of post hog, I get better insights out of Fable than 56 by quite a bit. multi-day Fable than 56 by quite a bit. multi-day Fable than 56 by quite a bit. multi-day runs. Soul is really, really good at. I runs. Soul is really, really good at. I runs. Soul is really, really good at. I wish I could say I tested Fable for wish I could say I tested Fable for wish I could say I tested Fable for these types of long multi-day runs, too, these types of long multi-day runs, too, these types of long multi-day runs, too, but that's just like absurdly expensive.

but that's just like absurdly expensive. but that's just like absurdly expensive. Not realistic, if I'm being honest with Not realistic, if I'm being honest with Not realistic, if I'm being honest with you guys. Now, we have the token you guys. Now, we have the token you guys. Now, we have the token efficiency side because it's so much efficiency side because it's so much efficiency side because it's so much more token efficient and faster. Yep. more token efficient and faster. Yep. more token efficient and faster. Yep. Said the same. Very good. The multi-day Said the same. Very good. The multi-day Said the same. Very good. The multi-day runs thing is crazy though, especially runs thing is crazy though, especially runs thing is crazy though, especially with the compaction. Buy four's with the compaction. Buy four's with the compaction. Buy four's compaction was great. Buy five broke it. compaction was great. Buy five broke it. compaction was great. Buy five broke it. So, when you had long threads, they So, when you had long threads, they So, when you had long threads, they would just lose track of what they're would just lose track of what they're would just lose track of what they're doing after it hit that point. 56 fully doing after it hit that point. 56 fully doing after it hit that point. 56 fully fixed it and I trust it to just go and fixed it and I trust it to just go and fixed it and I trust it to just go and go. I don't even look at the token usage go. I don't even look at the token usage go. I don't even look at the token usage anymore because it will compact and anymore because it will compact and anymore because it will compact and figure it [ __ ] out and then keep going figure it [ __ ] out and then keep going figure it [ __ ] out and then keep going and be fine. It's great. The downside is and be fine. It's great. The downside is and be fine. It's great. The downside is that you can feel Fable is naturally that you can feel Fable is naturally that you can feel Fable is naturally smarter. He had some brief moments of 56 smarter. He had some brief moments of 56 smarter. He had some brief moments of 56 where he was getting it to make fairly where he was getting it to make fairly where he was getting it to make fairly simple changes in eight turns. It just simple changes in eight turns. It just simple changes in eight turns. It just got stuck in dumb streams that was hard got stuck in dumb streams that was hard got stuck in dumb streams that was hard to get out of. So, while it's not AGI, to get out of. So, while it's not AGI, to get out of. So, while it's not AGI, don't get too carried away by the hype. don't get too carried away by the hype. don't get too carried away by the hype. I know this video has already been a bit I know this video has already been a bit I know this video has already been a bit long, but the wrap-up is where it's long, but the wrap-up is where it's long, but the wrap-up is where it's going to get really fun. I'm going to going to get really fun. I'm going to going to get really fun. I'm going to give you my exact methodology for how I give you my exact methodology for how I give you my exact methodology for how I pick which one to use and when, as well pick which one to use and when, as well pick which one to use and when, as well as the comparisons, the models generated as the comparisons, the models generated as the comparisons, the models generated of my actual long histories with both of of my actual long histories with both of of my actual long histories with both of them, so we can see how they feel about them, so we can see how they feel about them, so we can see how they feel about each other, too. So, I have these two each other, too. So, I have these two each other, too. So, I have these two reports that I had Fable and 56 generate reports that I had Fable and 56 generate reports that I had Fable and 56 generate based on my histories. I don't remember based on my histories. I don't remember based on my histories. I don't remember which one was generated by which. So, which one was generated by which. So, which one was generated by which. So, I'm going to use that CCF binding I I'm going to use that CCF binding I I'm going to use that CCF binding I mentioned before and see how quickly mentioned before and see how quickly mentioned before and see how quickly Soul could find it. And in under a Soul could find it. And in under a Soul could find it. And in under a minute and a half, probably it found minute and a half, probably it found minute and a half, probably it found both. So the OD one is quad code and the both. So the OD one is quad code and the both. So the OD one is quad code and the other one is codeex. Interesting. The other one is codeex. Interesting. The other one is codeex. Interesting. The pretty one was generated by codeex in 56 pretty one was generated by codeex in 56 pretty one was generated by codeex in 56 and the ugly one was believe it or not and the ugly one was believe it or not and the ugly one was believe it or not cla code. So let's read through these cla code. So let's read through these cla code. So let's read through these super quick and then I will give you my

super quick and then I will give you my super quick and then I will give you my mental model for how I pick which one to mental model for how I pick which one to mental model for how I pick which one to use and what I think you should use and what I think you should use and what I think you should subscribe to. Fable versus 56 soul. The subscribe to. Fable versus 56 soul. The subscribe to. Fable versus 56 soul. The fleet view. This again I use these fleet view. This again I use these fleet view. This again I use these across all my different machines. There across all my different machines. There across all my different machines. There were many of them and it collected all were many of them and it collected all were many of them and it collected all of the histories from all of them and of the histories from all of them and of the histories from all of them and analyzed it. According to Fable, the analyzed it. According to Fable, the analyzed it. According to Fable, the TLDDR is that 56 Soul is a distributed TLDDR is that 56 Soul is a distributed TLDDR is that 56 Soul is a distributed construction crew. Fable is a scarce construction crew. Fable is a scarce construction crew. Fable is a scarce senior engineer at two desks. I did a senior engineer at two desks. I did a senior engineer at two desks. I did a lot more work with 56 Soul. As I lot more work with 56 Soul. As I lot more work with 56 Soul. As I mentioned, they did not restrict me mentioned, they did not restrict me mentioned, they did not restrict me during early access testing, which during early access testing, which during early access testing, which they're probably going to in the future they're probably going to in the future they're probably going to in the future with how heavily I abused it. But as it with how heavily I abused it. But as it with how heavily I abused it. But as it said here, 87% of Soul's turns were said here, 87% of Soul's turns were said here, 87% of Soul's turns were machine delegated or scripted. So Soul machine delegated or scripted. So Soul machine delegated or scripted. So Soul was spinning up other sub agents. And was spinning up other sub agents. And was spinning up other sub agents. And 90% of the output I did was in a single 90% of the output I did was in a single 90% of the output I did was in a single 9-day swarm campaign where I was running 9-day swarm campaign where I was running 9-day swarm campaign where I was running a bunch of crazy rewrites on different a bunch of crazy rewrites on different a bunch of crazy rewrites on different machines. And also Fable was not around machines. And also Fable was not around machines. And also Fable was not around at that point. Fable lives where Theo at that point. Fable lives where Theo at that point. Fable lives where Theo physically sits. The MacBook and my physically sits. The MacBook and my physically sits. The MacBook and my framework desktop had 87% of the output. framework desktop had 87% of the output. framework desktop had 87% of the output. It does the planning before the swarms It does the planning before the swarms It does the planning before the swarms and then judges after. And on the six and then judges after. And on the six and then judges after. And on the six days both models were active, Fable days both models were active, Fable days both models were active, Fable outproduced Soul by 1.3x roughly. So outproduced Soul by 1.3x roughly. So outproduced Soul by 1.3x roughly. So again, on the days I'm actually using again, on the days I'm actually using again, on the days I'm actually using the models and I have access to both, I the models and I have access to both, I the models and I have access to both, I definitely lean Fable. It didn't really definitely lean Fable. It didn't really definitely lean Fable. It didn't really get what I was asking for though. I get what I was asking for though. I get what I was asking for though. I wanted to compare more how the models wanted to compare more how the models wanted to compare more how the models behave. Thankfully, Soul understood that behave. Thankfully, Soul understood that behave. Thankfully, Soul understood that because it takes me more literally. So, because it takes me more literally. So, because it takes me more literally. So, Soul wrote up a report about how they Soul wrote up a report about how they Soul wrote up a report about how they behave. It said the Fable leaned behave. It said the Fable leaned behave. It said the Fable leaned collaborator and 56 leaned system. Both collaborator and 56 leaned system. Both collaborator and 56 leaned system. Both did direct work in orchestration. Fable did direct work in orchestration. Fable did direct work in orchestration. Fable was more often the frontline was more often the frontline was more often the frontline collaborator and 56 was more often a collaborator and 56 was more often a collaborator and 56 was more often a persistent orchestration layer. 56

persistent orchestration layer. 56 persistent orchestration layer. 56 produced far more recorded token volume, produced far more recorded token volume, produced far more recorded token volume, but almost all of the gap came from a but almost all of the gap came from a but almost all of the gap came from a 9-day automationheavy campaign. Fable 9-day automationheavy campaign. Fable 9-day automationheavy campaign. Fable appears in smaller direct product appears in smaller direct product appears in smaller direct product building bursts. I love that here it building bursts. I love that here it building bursts. I love that here it rendered these charts and you can rendered these charts and you can rendered these charts and you can clearly see when I had to stop using clearly see when I had to stop using clearly see when I had to stop using each model and different machines had each model and different machines had each model and different machines had different stories. Fable was mostly on different stories. Fable was mostly on different stories. Fable was mostly on my MacBook. 56 was on the Zbook which is my MacBook. 56 was on the Zbook which is my MacBook. 56 was on the Zbook which is the computer I had doing those crazy the computer I had doing those crazy the computer I had doing those crazy rewrites I showed earlier. I gave them rewrites I showed earlier. I gave them rewrites I showed earlier. I gave them different jobs where Fable was a different jobs where Fable was a different jobs where Fable was a frontline collaborator and 56 was an frontline collaborator and 56 was an frontline collaborator and 56 was an orchestration layer. On comparable orchestration layer. On comparable orchestration layer. On comparable dates, the Gulf nearly disappears. What dates, the Gulf nearly disappears. What dates, the Gulf nearly disappears. What this report cannot tell you is quality. this report cannot tell you is quality. this report cannot tell you is quality. It's not actually measuring the outputs. It's not actually measuring the outputs. It's not actually measuring the outputs. Yeah, still cool to see it do a nice Yeah, still cool to see it do a nice Yeah, still cool to see it do a nice fancy report like this, but I wish it fancy report like this, but I wish it fancy report like this, but I wish it did more like for like comparisons did more like for like comparisons did more like for like comparisons between Fable and 56. I might ask it to between Fable and 56. I might ask it to between Fable and 56. I might ask it to do another set of reports, but I don't do another set of reports, but I don't do another set of reports, but I don't want to wait to do all of that now. I'll want to wait to do all of that now. I'll want to wait to do all of that now. I'll include that in future videos if I get include that in future videos if I get include that in future videos if I get anything good because right now I want anything good because right now I want anything good because right now I want to use your time efficiently and give to use your time efficiently and give to use your time efficiently and give you the most important thing, when to you the most important thing, when to you the most important thing, when to use each, but more importantly, which to use each, but more importantly, which to use each, but more importantly, which to subscribe to. I have used a lot of all subscribe to. I have used a lot of all subscribe to. I have used a lot of all of these subs and I've used a lot of of these subs and I've used a lot of of these subs and I've used a lot of both of these models which is why my both of these models which is why my both of these models which is why my response here might surprise you. I response here might surprise you. I response here might surprise you. I prefer Fable and I prefer Fable for a prefer Fable and I prefer Fable for a prefer Fable and I prefer Fable for a lot of things. Specifically, I use Fable lot of things. Specifically, I use Fable lot of things. Specifically, I use Fable if I want the change to land. I use if I want the change to land. I use if I want the change to land. I use Fable if I want a good design. I use Fable if I want a good design. I use Fable if I want a good design. I use Fable if the problem is hard or complex Fable if the problem is hard or complex Fable if the problem is hard or complex in weird ways. I use Fable to figure out in weird ways. I use Fable to figure out in weird ways. I use Fable to figure out what I want. And I use Fable to verify what I want. And I use Fable to verify what I want. And I use Fable to verify and simplify work from other things,

and simplify work from other things, and simplify work from other things, even itself. And honestly, the first one even itself. And honestly, the first one even itself. And honestly, the first one should be enough to convince you that should be enough to convince you that should be enough to convince you that Fable is an incredible model. So, what Fable is an incredible model. So, what Fable is an incredible model. So, what the hell do I use 56 for then? Well, I the hell do I use 56 for then? Well, I the hell do I use 56 for then? Well, I use 56 first almost always because it's use 56 first almost always because it's use 56 first almost always because it's so much cheaper. It's practically free so much cheaper. It's practically free so much cheaper. It's practically free and if it turns out 56 can solve it and and if it turns out 56 can solve it and and if it turns out 56 can solve it and it can save me all the token and usage it can save me all the token and usage it can save me all the token and usage burn from Fable, it's worth it. So, I burn from Fable, it's worth it. So, I burn from Fable, it's worth it. So, I always start with 56 for the work that I always start with 56 for the work that I always start with 56 for the work that I would do in either just to see where it would do in either just to see where it would do in either just to see where it goes. I use it for long running work all goes. I use it for long running work all goes. I use it for long running work all the time because it's so much better at the time because it's so much better at the time because it's so much better at staying on task for these runs that go staying on task for these runs that go staying on task for these runs that go for hours and hours. It's genuinely for hours and hours. It's genuinely for hours and hours. It's genuinely really impressive. I've had it run for really impressive. I've had it run for really impressive. I've had it run for days at a time and actually make working days at a time and actually make working days at a time and actually make working outputs, which is still crazy to me outputs, which is still crazy to me outputs, which is still crazy to me conceptually, but it works. It does it. conceptually, but it works. It does it. conceptually, but it works. It does it. It blows me away. The fact that I can It blows me away. The fact that I can It blows me away. The fact that I can have it on a goal and it goes for two have it on a goal and it goes for two have it on a goal and it goes for two and a half days and then stops and says, and a half days and then stops and says, and a half days and then stops and says, "Okay, I did it. Here's what I built. "Okay, I did it. Here's what I built. "Okay, I did it. Here's what I built. Here's what I want you to go test. And Here's what I want you to go test. And Here's what I want you to go test. And if depending on how this goes, I can if depending on how this goes, I can if depending on how this goes, I can continue." Really incredible. and it continue." Really incredible. and it continue." Really incredible. and it sends me things that work after days. sends me things that work after days. sends me things that work after days. And since it's so much cheaper, this is And since it's so much cheaper, this is And since it's so much cheaper, this is actually somewhat reasonable, unlike on actually somewhat reasonable, unlike on actually somewhat reasonable, unlike on other models. I absolutely adore 56 for other models. I absolutely adore 56 for other models. I absolutely adore 56 for computer use. I use it way more than I computer use. I use it way more than I computer use. I use it way more than I ever thought I would. I did personally ever thought I would. I did personally ever thought I would. I did personally just didn't like computer use before to just didn't like computer use before to just didn't like computer use before to the point where I would like talk [ __ ] the point where I would like talk [ __ ] the point where I would like talk [ __ ] on it, and I've intentionally not on it, and I've intentionally not on it, and I've intentionally not invested in companies building in that invested in companies building in that invested in companies building in that space. I regret that now because the space. I regret that now because the space. I regret that now because the models, at least the OpenAI ones, have models, at least the OpenAI ones, have models, at least the OpenAI ones, have gotten really, really good at it, and it gotten really, really good at it, and it gotten really, really good at it, and it blows me away what they're capable of. I blows me away what they're capable of. I blows me away what they're capable of. I have lots of examples of that in my 56 have lots of examples of that in my 56 have lots of examples of that in my 56 review video. You should definitely review video. You should definitely review video. You should definitely check that if you haven't. It's melted check that if you haven't. It's melted check that if you haven't. It's melted my brain. On the note of like general my brain. On the note of like general my brain. On the note of like general use, 56 in tools like Hermes and use, 56 in tools like Hermes and use, 56 in tools like Hermes and OpenClaw is incredible. Both because you

OpenClaw is incredible. Both because you OpenClaw is incredible. Both because you get the crazy subsidization with the get the crazy subsidization with the get the crazy subsidization with the subscription, but also because it takes subscription, but also because it takes subscription, but also because it takes intent literally and responds quickly. intent literally and responds quickly. intent literally and responds quickly. So, I really like it for those use So, I really like it for those use So, I really like it for those use cases, too. As well as like quick fixes cases, too. As well as like quick fixes cases, too. As well as like quick fixes in general. Whether the quick fix is in general. Whether the quick fix is in general. Whether the quick fix is something on my computer, in my system something on my computer, in my system something on my computer, in my system maintenance, or if the fix is something maintenance, or if the fix is something maintenance, or if the fix is something that is in a codebase that should be that is in a codebase that should be that is in a codebase that should be really quick. I love using 56 for oneoff really quick. I love using 56 for oneoff really quick. I love using 56 for oneoff things where I'm going to watch the things where I'm going to watch the things where I'm going to watch the terminal from the start to the finish. terminal from the start to the finish. terminal from the start to the finish. If I'm going to watch the terminal the If I'm going to watch the terminal the If I'm going to watch the terminal the whole time or I'm going to watch it in whole time or I'm going to watch it in whole time or I'm going to watch it in T3 code the whole time, 56 fast low is T3 code the whole time, 56 fast low is T3 code the whole time, 56 fast low is one of my favorite things to use. But if one of my favorite things to use. But if one of my favorite things to use. But if I'm going to open it in the background I'm going to open it in the background I'm going to open it in the background and then come back in 30 minutes to an and then come back in 30 minutes to an and then come back in 30 minutes to an hour and want to change, I can hit merge hour and want to change, I can hit merge hour and want to change, I can hit merge on. I still obviously prefer Fable. I on. I still obviously prefer Fable. I on. I still obviously prefer Fable. I also really like delegating to 56 to also really like delegating to 56 to also really like delegating to 56 to have another model like Fable break up have another model like Fable break up have another model like Fable break up work into smaller pieces and then tell work into smaller pieces and then tell work into smaller pieces and then tell 56 to go complete a part of it. It's a 56 to go complete a part of it. It's a 56 to go complete a part of it. It's a way to save meaningful amounts of usage way to save meaningful amounts of usage way to save meaningful amounts of usage while also getting really good results. while also getting really good results. while also getting really good results. And I've loved 56 for this and having And I've loved 56 for this and having And I've loved 56 for this and having Fable be an orchestrator. We showed this Fable be an orchestrator. We showed this Fable be an orchestrator. We showed this a lot in other content. Definitely check a lot in other content. Definitely check a lot in other content. Definitely check that all out. But I do really like 56 as that all out. But I do really like 56 as that all out. But I do really like 56 as a thing that is called on to do stuff.

a thing that is called on to do stuff. a thing that is called on to do stuff. The simplest way to put this is I kind The simplest way to put this is I kind The simplest way to put this is I kind of treat Fable like a really smart of treat Fable like a really smart of treat Fable like a really smart co-orker. And just because your co-orker co-orker. And just because your co-orker co-orker. And just because your co-orker is smart doesn't mean they can't do is smart doesn't mean they can't do is smart doesn't mean they can't do stupid things or have bad opinions stupid things or have bad opinions stupid things or have bad opinions sometimes. It just means that they're sometimes. It just means that they're sometimes. It just means that they're generally pretty smart and get [ __ ] generally pretty smart and get [ __ ] generally pretty smart and get [ __ ] Where with 56, 56 is a really strong Where with 56, 56 is a really strong Where with 56, 56 is a really strong tool. It is a tool that I love using and tool. It is a tool that I love using and tool. It is a tool that I love using and I wield for all sorts of different I wield for all sorts of different I wield for all sorts of different stuff, but they are very very different. stuff, but they are very very different. stuff, but they are very very different. Which is why it is crazy they bench so Which is why it is crazy they bench so Which is why it is crazy they bench so closely and they behave so to speak on closely and they behave so to speak on closely and they behave so to speak on benches as similarly as they do because benches as similarly as they do because benches as similarly as they do because using these models couldn't feel more using these models couldn't feel more using these models couldn't feel more different when I do it every day. And different when I do it every day. And different when I do it every day. And this is where the the hard part comes this is where the the hard part comes this is where the the hard part comes gun to my head. I can only pick one. gun to my head. I can only pick one. gun to my head. I can only pick one. What do I pick? This is actually really What do I pick? This is actually really What do I pick? This is actually really hard. I have thought about this so much hard. I have thought about this so much hard. I have thought about this so much both online and offline and it's both online and offline and it's both online and offline and it's difficult and I'm going to give a weird difficult and I'm going to give a weird difficult and I'm going to give a weird answer here. If I have to pick just answer here. If I have to pick just answer here. If I have to pick just between these two things as what I use between these two things as what I use between these two things as what I use for my work, my code and everything for my work, my code and everything for my work, my code and everything daytoday, daytoday, daytoday, Fable wins. But if I look at my own life Fable wins. But if I look at my own life Fable wins. But if I look at my own life and the things around me, this is and the things around me, this is and the things around me, this is reductive reductive reductive because I have Julius. I have one of the because I have Julius. I have one of the because I have Julius. I have one of the best engineers ever working with me, best engineers ever working with me, best engineers ever working with me, building awesome [ __ ] who has really building awesome [ __ ] who has really building awesome [ __ ] who has really good taste, who builds really competent good taste, who builds really competent good taste, who builds really competent stuff, who puts up code that can stuff, who puts up code that can stuff, who puts up code that can absolutely land. If I didn't have Julius absolutely land. If I didn't have Julius absolutely land. If I didn't have Julius on my team, I don't know how I would on my team, I don't know how I would on my team, I don't know how I would function without Fable. But because I function without Fable. But because I function without Fable. But because I have Julius and I can trust him to do have Julius and I can trust him to do have Julius and I can trust him to do that top level of work, I think I would

that top level of work, I think I would that top level of work, I think I would struggle more without 56 because the way struggle more without 56 because the way struggle more without 56 because the way I use Fable is honestly similar to the I use Fable is honestly similar to the I use Fable is honestly similar to the way I text Julius. I give a vague idea way I text Julius. I give a vague idea way I text Julius. I give a vague idea of a problem I have and a rough of a problem I have and a rough of a problem I have and a rough direction I want to go in, and Fable can direction I want to go in, and Fable can direction I want to go in, and Fable can mostly figure it out, as can Julius, mostly figure it out, as can Julius, mostly figure it out, as can Julius, even better than Fable can. most of the even better than Fable can. most of the even better than Fable can. most of the time. But when I just want to like do time. But when I just want to like do time. But when I just want to like do some random [ __ ] on my computer or fix some random [ __ ] on my computer or fix some random [ __ ] on my computer or fix something on another machine in my something on another machine in my something on another machine in my network or make a quick one-off fix, the network or make a quick one-off fix, the network or make a quick one-off fix, the type of thing that I wouldn't want to type of thing that I wouldn't want to type of thing that I wouldn't want to text to my coworker Julius, then I much text to my coworker Julius, then I much text to my coworker Julius, then I much prefer 56. And I am sorry chat, you guys prefer 56. And I am sorry chat, you guys prefer 56. And I am sorry chat, you guys don't have access to Julius. He is mine. don't have access to Julius. He is mine. don't have access to Julius. He is mine. And I would recommend you guys give him And I would recommend you guys give him And I would recommend you guys give him a follow if you want to see what a follow if you want to see what a follow if you want to see what building at this scale and at these building at this scale and at these building at this scale and at these levels looks like. He's unbelievably levels looks like. He's unbelievably levels looks like. He's unbelievably talented. And since I have a team with talented. And since I have a team with talented. And since I have a team with Julius and other similarly capable Julius and other similarly capable Julius and other similarly capable engineers coming on too, my need for a engineers coming on too, my need for a engineers coming on too, my need for a autonomous model that can clean up the autonomous model that can clean up the autonomous model that can clean up the code itself and land the change is less code itself and land the change is less code itself and land the change is less great than it was without those things.

great than it was without those things. great than it was without those things. If I was a solo dev or a solo founder, I If I was a solo dev or a solo founder, I If I was a solo dev or a solo founder, I would rely on Fable for everything I do. would rely on Fable for everything I do. would rely on Fable for everything I do. But since I'm an engineer at a company But since I'm an engineer at a company But since I'm an engineer at a company with other engineers who are incredibly with other engineers who are incredibly with other engineers who are incredibly capable, too, 56 feels more like a capable, too, 56 feels more like a capable, too, 56 feels more like a hammer in my toolbox that I reach for hammer in my toolbox that I reach for hammer in my toolbox that I reach for all the time. Fable feels like the all the time. Fable feels like the all the time. Fable feels like the contractor that I hire to come in and contractor that I hire to come in and contractor that I hire to come in and use their tools themselves. So, with use their tools themselves. So, with use their tools themselves. So, with that, which sub should you buy? I will that, which sub should you buy? I will that, which sub should you buy? I will say very confidently, every dev who can say very confidently, every dev who can say very confidently, every dev who can afford it should have at least the $100 afford it should have at least the $100 afford it should have at least the $100 tier for Codeex. It is just so absurdly tier for Codeex. It is just so absurdly tier for Codeex. It is just so absurdly generous, especially with the constant generous, especially with the constant generous, especially with the constant resets that we've been getting as of resets that we've been getting as of resets that we've been getting as of late. You should have that $100 tier. late. You should have that $100 tier. late. You should have that $100 tier. You should find ways to max it out that You should find ways to max it out that You should find ways to max it out that bring actual value. Whether that is a bring actual value. Whether that is a bring actual value. Whether that is a Hermes agent or a cron job that indexes Hermes agent or a cron job that indexes Hermes agent or a cron job that indexes your PRs every day or you start going your PRs every day or you start going your PRs every day or you start going for bolder work, whatever it is, you for bolder work, whatever it is, you for bolder work, whatever it is, you should not make a decision about which should not make a decision about which should not make a decision about which subscription to use until you have maxed subscription to use until you have maxed subscription to use until you have maxed out a $100 tier on codecs. And then you out a $100 tier on codecs. And then you out a $100 tier on codecs. And then you could sit and look at the work you've could sit and look at the work you've could sit and look at the work you've done and decide, is this useful? Do I done and decide, is this useful? Do I done and decide, is this useful? Do I like the outputs? Can I see myself using like the outputs? Can I see myself using like the outputs? Can I see myself using this even more? If so, upgrade that to this even more? If so, upgrade that to this even more? If so, upgrade that to the $200 codeex. If not, go open up a the $200 codeex. If not, go open up a the $200 codeex. If not, go open up a $100 Claude code and see how you like $100 Claude code and see how you like $100 Claude code and see how you like the results from Fable. I'll be honest the results from Fable. I'll be honest the results from Fable. I'll be honest though, I would much more quickly cancel though, I would much more quickly cancel though, I would much more quickly cancel my Claude code sub than my codec sub, as my Claude code sub than my codec sub, as my Claude code sub than my codec sub, as I've made clear in the past, because I I've made clear in the past, because I I've made clear in the past, because I don't know when and how I can use that don't know when and how I can use that don't know when and how I can use that sub and all the things I can use it for.

sub and all the things I can use it for. sub and all the things I can use it for. If you're not hitting the limits on your If you're not hitting the limits on your If you're not hitting the limits on your codec sub, you're not getting creative codec sub, you're not getting creative codec sub, you're not getting creative enough. And you can do some crazy things enough. And you can do some crazy things enough. And you can do some crazy things with that subscription, which is why I with that subscription, which is why I with that subscription, which is why I love mine so much and why I've been love mine so much and why I've been love mine so much and why I've been getting crazier at hitting those rate getting crazier at hitting those rate getting crazier at hitting those rate limits as of recent because I'm using it limits as of recent because I'm using it limits as of recent because I'm using it for all sorts of different things. It's for all sorts of different things. It's for all sorts of different things. It's just so valuable a tool to have in your just so valuable a tool to have in your just so valuable a tool to have in your toolbox that I think you should have toolbox that I think you should have toolbox that I think you should have that first and get used to using it that first and get used to using it that first and get used to using it first before you look at alternatives to first before you look at alternatives to first before you look at alternatives to toolboxes. And of course, there's the toolboxes. And of course, there's the toolboxes. And of course, there's the problem that Fable is going to be problem that Fable is going to be problem that Fable is going to be removed from the subscription tier soon. removed from the subscription tier soon. removed from the subscription tier soon. It might not even be there by the time It might not even be there by the time It might not even be there by the time you're watching this video. Will it come you're watching this video. Will it come you're watching this video. Will it come back eventually? Yeah, who knows when back eventually? Yeah, who knows when back eventually? Yeah, who knows when though? So for all of these reasons, I though? So for all of these reasons, I though? So for all of these reasons, I think everyone should start with the think everyone should start with the think everyone should start with the codec subscription to use it to push its codec subscription to use it to push its codec subscription to use it to push its limits to get as much out of it as they limits to get as much out of it as they limits to get as much out of it as they can. And if you hate the codec and you can. And if you hate the codec and you can. And if you hate the codec and you don't like the chat GBT app, I don't like the chat GBT app, I don't like the chat GBT app, I understand. I recommend T3 code for an understand. I recommend T3 code for an understand. I recommend T3 code for an alternative to the chat GBT app. Or you alternative to the chat GBT app. Or you alternative to the chat GBT app. Or you can watch my video all about using your can watch my video all about using your can watch my video all about using your codec sub include code in order to get codec sub include code in order to get codec sub include code in order to get the benefits of that harness. There are the benefits of that harness. There are the benefits of that harness. There are some rough edges there, but I found it some rough edges there, but I found it some rough edges there, but I found it to be a very pleasant experience. And to be a very pleasant experience. And to be a very pleasant experience. And with that, I think I'm finally done with with that, I think I'm finally done with with that, I think I'm finally done with this topic. I have my answer to the this topic. I have my answer to the this topic. I have my answer to the question everyone keeps asking me, which question everyone keeps asking me, which question everyone keeps asking me, which is which model is smarter and better, is which model is smarter and better, is which model is smarter and better, and which one do I prefer for different and which one do I prefer for different and which one do I prefer for different things. I've said all of it now across things. I've said all of it now across things. I've said all of it now across all this different content, but this all this different content, but this all this different content, but this video in particular, I hope it is clear video in particular, I hope it is clear video in particular, I hope it is clear why and how I choose each model, and the why and how I choose each model, and the why and how I choose each model, and the simplest way I can put it is I use 56 simplest way I can put it is I use 56 simplest way I can put it is I use 56 most of the time, and if it doesn't do most of the time, and if it doesn't do most of the time, and if it doesn't do what I want it to, I go over to Fable to what I want it to, I go over to Fable to what I want it to, I go over to Fable to get a good change that I can actually get a good change that I can actually get a good change that I can actually get merged. It's truly incredible to get merged. It's truly incredible to get merged. It's truly incredible to have two options of this absurd level of

have two options of this absurd level of have two options of this absurd level of quality. And for them to be so different quality. And for them to be so different quality. And for them to be so different is just fascinating and fun. It's never is just fascinating and fun. It's never is just fascinating and fun. It's never been more enjoyable to play with models been more enjoyable to play with models been more enjoyable to play with models the way it is today. And I've had so the way it is today. And I've had so the way it is today. And I've had so much fun experimenting with both Fable 5 much fun experimenting with both Fable 5 much fun experimenting with both Fable 5 and GPT56 Soul that I recommend you do and GPT56 Soul that I recommend you do and GPT56 Soul that I recommend you do the same. Don't just blindly follow my the same. Don't just blindly follow my the same. Don't just blindly follow my recommendations here. If you take what I recommendations here. If you take what I recommendations here. If you take what I did and then copy paste it into your did and then copy paste it into your did and then copy paste it into your setup because obviously I know it's setup because obviously I know it's setup because obviously I know it's right. I'm the influencer. Don't take right. I'm the influencer. Don't take right. I'm the influencer. Don't take lessons from what I did, learn from how lessons from what I did, learn from how lessons from what I did, learn from how I think about these things and then go I think about these things and then go I think about these things and then go try them yourself. Maybe you agree fully try them yourself. Maybe you agree fully try them yourself. Maybe you agree fully with what I did and what I say. Maybe with what I did and what I say. Maybe with what I did and what I say. Maybe you end up going in an entirely you end up going in an entirely you end up going in an entirely different direction and you feel different direction and you feel different direction and you feel fundamentally different than I do about fundamentally different than I do about fundamentally different than I do about all of this. I can't know. That's not my all of this. I can't know. That's not my all of this. I can't know. That's not my job. That's your job. Now, experimenting job. That's your job. Now, experimenting job. That's your job. Now, experimenting with these things is one of the best with these things is one of the best with these things is one of the best value ads you can have as an engineer. value ads you can have as an engineer. value ads you can have as an engineer. And then bringing that knowledge to your And then bringing that knowledge to your And then bringing that knowledge to your teams or other companies will help you teams or other companies will help you teams or other companies will help you level up more and more. So, while this level up more and more. So, while this level up more and more. So, while this video is definitive for me, I hope that video is definitive for me, I hope that video is definitive for me, I hope that it's just the start for you. Let this it's just the start for you. Let this it's just the start for you. Let this open up the journey to go play with open up the journey to go play with open up the journey to go play with these things more. Experiment with the these things more. Experiment with the these things more. Experiment with the models and the tools and the harnesses models and the tools and the harnesses models and the tools and the harnesses you use them in. change the types of you use them in. change the types of you use them in. change the types of tasks and work you use them for and see tasks and work you use them for and see tasks and work you use them for and see how far you can push them. I have a how far you can push them. I have a how far you can push them. I have a feeling you'll be surprised just like I feeling you'll be surprised just like I feeling you'll be surprised just like I was. That's all I have to say about was. That's all I have to say about was. That's all I have to say about this. So until next time, these nerds