← Back
Theo July 10, 2026 26m

So I've been using gpt-5.6 for awhile...

Read full transcript 21 segments
  1. GPT 5.6 is finally here, and I don't GPT 5.6 is finally here, and I don't think I've ever been this excited to think I've ever been this excited to think I've ever been this excited to talk about a model. There's a reason talk about a model. There's a reason talk about a model. There's a reason why, though. It's not cuz it's the best why, though. It's not cuz it's the best why, though. It's not cuz it's the best thing ever. It is really great, but the thing ever. It is really great, but the thing ever. It is really great, but the reason is cuz I've had access for a reason is cuz I've had access for a reason is cuz I've had access for a while now, since before Fable even first while now, since before Fable even first while now, since before Fable even first dropped. I've been waiting to talk about dropped. I've been waiting to talk about dropped. I've been waiting to talk about this model for so goddamn long that I've this model for so goddamn long that I've this model for so goddamn long that I've been putting out less videos cuz I've been putting out less videos cuz I've been putting out less videos cuz I've been trying to clean up my computer to been trying to clean up my computer to been trying to clean up my computer to hide the fact that I've used it so much. hide the fact that I've used it so much. hide the fact that I've used it so much. And when I say so much, I mean it. And when I say so much, I mean it. And when I say so much, I mean it. Depending on how you estimate my usage, Depending on how you estimate my usage, Depending on how you estimate my usage, I did between $180,000 and $240,000 I did between $180,000 and $240,000 I did between $180,000 and $240,000 of inference in my month and a half-ish of inference in my month and a half-ish of inference in my month and a half-ish of access. And I got an absurd amount of access. And I got an absurd amount of access. And I got an absurd amount done in that time. It took a bit for me done in that time. It took a bit for me done in that time. It took a bit for me to digest what made 5.6 so special, but to digest what made 5.6 so special, but to digest what made 5.6 so special, but once I did, I started pushing it further once I did, I started pushing it further once I did, I started pushing it further than I've ever pushed a model. I did than I've ever pushed a model. I did than I've ever pushed a model. I did more inference in the past month than I more inference in the past month than I more inference in the past month than I had done in my life up until that point. had done in my life up until that point. had done in my life up until that point. And I have so many thoughts to share now And I have so many thoughts to share now And I have so many thoughts to share now that I've done this. I need to be that I've done this. I need to be that I've done this. I need to be realistic, though. If I share everything realistic, though. If I share everything realistic, though. If I share everything I have to say about this model, the I have to say about this model, the I have to say about this model, the video is going to be hours upon hours video is going to be hours upon hours video is going to be hours upon hours long. So, instead, I'm going to break long. So, instead, I'm going to break long. So, instead, I'm going to break this up into multiple different parts. this up into multiple different parts. this up into multiple different parts. And this one's particularly special. The And this one's particularly special. The And this one's particularly special. The reason this one's special is cuz I reason this one's special is cuz I reason this one's special is cuz I haven't read any of the coverage on this haven't read any of the coverage on this haven't read any of the coverage on this model yet. I have no idea what people model yet. I have no idea what people model yet. I have no idea what people think about it, how they're using it, or think about it, how they're using it, or think about it, how they're using it, or what they expect from a video like this.

  2. what they expect from a video like this. what they expect from a video like this. I'm just going off the top. I'm sharing I'm just going off the top. I'm sharing I'm just going off the top. I'm sharing how I used the model and how I spent how I used the model and how I spent how I used the model and how I spent such an absurd amount of money on this such an absurd amount of money on this such an absurd amount of money on this model to show you guys what it's model to show you guys what it's model to show you guys what it's actually like, or at least what it was actually like, or at least what it was actually like, or at least what it was like for me. What I'm trying to say here like for me. What I'm trying to say here like for me. What I'm trying to say here is don't think of this video as a is don't think of this video as a is don't think of this video as a review. I'll have a proper 5.6 review review. I'll have a proper 5.6 review review. I'll have a proper 5.6 review coming up really soon after this. Right coming up really soon after this. Right coming up really soon after this. Right now, I just want to show how I used the now, I just want to show how I used the now, I just want to show how I used the model and the cool things I built with model and the cool things I built with model and the cool things I built with it cuz turns out you can do a lot with it cuz turns out you can do a lot with it cuz turns out you can do a lot with $200,000 of tokens. What that means is I $200,000 of tokens. What that means is I $200,000 of tokens. What that means is I won't be talking about Luna and Terra won't be talking about Luna and Terra won't be talking about Luna and Terra cuz I didn't have access during the cuz I didn't have access during the cuz I didn't have access during the testing. I won't be comparing with testing. I won't be comparing with testing. I won't be comparing with Fable. That has its own dedicated video. Fable. That has its own dedicated video. Fable. That has its own dedicated video. I won't talk about ultra or max I won't talk about ultra or max I won't talk about ultra or max reasoning levels because one of those is reasoning levels because one of those is reasoning levels because one of those is a reasoning level, the others are new a reasoning level, the others are new a reasoning level, the others are new feature, neither of which are things I feature, neither of which are things I feature, neither of which are things I had access to. I won't talk about the had access to. I won't talk about the had access to. I won't talk about the death of Codex even though, believe me, death of Codex even though, believe me, death of Codex even though, believe me, I want to talk about this. I have a lot I want to talk about this. I have a lot I want to talk about this. I have a lot of things to say. I won't talk about of things to say. I won't talk about of things to say. I won't talk about benchmarks or sentiment cuz again, I benchmarks or sentiment cuz again, I benchmarks or sentiment cuz again, I haven't seen any of them other than what haven't seen any of them other than what haven't seen any of them other than what was posted with that initial was posted with that initial was posted with that initial announcement, and those benches didn't announcement, and those benches didn't announcement, and those benches didn't really make sense to me. I won't talk really make sense to me. I won't talk really make sense to me. I won't talk about pricing or usage limits, because about pricing or usage limits, because about pricing or usage limits, because again, I didn't have any of that when I again, I didn't have any of that when I again, I didn't have any of that when I was testing the model, and I won't talk was testing the model, and I won't talk was testing the model, and I won't talk about the fast version, because it about the fast version, because it about the fast version, because it didn't exist. My fast version was 1.5x didn't exist. My fast version was 1.5x didn't exist. My fast version was 1.5x faster, not the 750 TPS that's being faster, not the 750 TPS that's being faster, not the 750 TPS that's being promised by Cerebras. I don't even know promised by Cerebras. I don't even know promised by Cerebras. I don't even know if that's out yet. Again, I haven't if that's out yet. Again, I haven't if that's out yet. Again, I haven't checked anything today. What I have checked anything today. What I have checked anything today. What I have checked out though, is today's sponsor.

  3. checked out though, is today's sponsor. checked out though, is today's sponsor. I'm going to be real with you guys, I I'm going to be real with you guys, I I'm going to be real with you guys, I was super, super skeptical of computer was super, super skeptical of computer was super, super skeptical of computer use and browser use when I first saw it use and browser use when I first saw it use and browser use when I first saw it popping up. It just didn't make sense to popping up. It just didn't make sense to popping up. It just didn't make sense to me. Why would an agent be faster than a me. Why would an agent be faster than a me. Why would an agent be faster than a human clicking buttons, if it has to human clicking buttons, if it has to human clicking buttons, if it has to screenshot, manually move the mouse to a screenshot, manually move the mouse to a screenshot, manually move the mouse to a specific coordinate, click, and then specific coordinate, click, and then specific coordinate, click, and then wait, and hope it gets things right? And wait, and hope it gets things right? And wait, and hope it gets things right? And my early experiences with computer use my early experiences with computer use my early experiences with computer use stuff showed just how bad it was, and I stuff showed just how bad it was, and I stuff showed just how bad it was, and I moved on. But then something important moved on. But then something important moved on. But then something important happened. GPT-5.6. happened. GPT-5.6. happened. GPT-5.6. There's a lot of things that make this There's a lot of things that make this There's a lot of things that make this model special, but the one that model special, but the one that model special, but the one that impressed me the most immediately was impressed me the most immediately was impressed me the most immediately was how much better it was at using a how much better it was at using a how much better it was at using a computer, in particular using browsers. computer, in particular using browsers. computer, in particular using browsers. Like you can put a task like, set this Like you can put a task like, set this Like you can put a task like, set this up in a dashboard, make this change in up in a dashboard, make this change in up in a dashboard, make this change in Notion, and just navigate the web for Notion, and just navigate the web for Notion, and just navigate the web for me, and it would. But then I had to me, and it would. But then I had to me, and it would. But then I had to dedicate my whole computer to letting dedicate my whole computer to letting dedicate my whole computer to letting CodeX use it. Well, I did, before I set CodeX use it. Well, I did, before I set CodeX use it. Well, I did, before I set it browser-based. Browser-based built it browser-based. Browser-based built it browser-based. Browser-based built the browser for your agents. They allow the browser for your agents. They allow the browser for your agents. They allow for programmatic access to the entire for programmatic access to the entire for programmatic access to the entire web through their Puppeteer layer that web through their Puppeteer layer that web through their Puppeteer layer that they are hosting for you. You no longer they are hosting for you. You no longer they are hosting for you. You no longer have to deal with Playwright yourself. have to deal with Playwright yourself. have to deal with Playwright yourself. You don't have to set up a browser or a You don't have to set up a browser or a You don't have to set up a browser or a real computer, you just give an API to real computer, you just give an API to real computer, you just give an API to your agents, and now they can do your agents, and now they can do your agents, and now they can do whatever they need to on the web. And whatever they need to on the web. And whatever they need to on the web. And this isn't just some small startup guys.

  4. this isn't just some small startup guys. this isn't just some small startup guys. Everyone from Microsoft to Lovable to Everyone from Microsoft to Lovable to Everyone from Microsoft to Lovable to Amplitude is already relying on Amplitude is already relying on Amplitude is already relying on Browser-based for running their Browser-based for running their Browser-based for running their browsers. Even DeepMind has been using browsers. Even DeepMind has been using browsers. Even DeepMind has been using them recently, which is just them recently, which is just them recently, which is just unbelievable. But setting up browsers unbelievable. But setting up browsers unbelievable. But setting up browsers for your agents isn't easy. And in a for your agents isn't easy. And in a for your agents isn't easy. And in a world where 85% of the web's APIs are world where 85% of the web's APIs are world where 85% of the web's APIs are blocked behind UI, a browser becomes blocked behind UI, a browser becomes blocked behind UI, a browser becomes essential for doing real work. Give your essential for doing real work. Give your essential for doing real work. Give your agents the entire web at agents the entire web at agents the entire web at soda.link/browserbase. soda.link/browserbase. soda.link/browserbase. A few more things I want to clarify A few more things I want to clarify A few more things I want to clarify before we dive too deep. OpenAI has not before we dive too deep. OpenAI has not before we dive too deep. OpenAI has not paid me for any of this coverage. paid me for any of this coverage. paid me for any of this coverage. Normally, I don't do this level of usage Normally, I don't do this level of usage Normally, I don't do this level of usage when they give me early access. I do when they give me early access. I do when they give me early access. I do maybe a hundred or two hundred dollars maybe a hundred or two hundred dollars maybe a hundred or two hundred dollars of inference, which is covered under my of inference, which is covered under my of inference, which is covered under my paid two hundred dollar plan anyways, paid two hundred dollar plan anyways, paid two hundred dollar plan anyways, which I'm still paying full price for. which I'm still paying full price for. which I'm still paying full price for. In fact, I have a bad feeling I might In fact, I have a bad feeling I might In fact, I have a bad feeling I might have to open up another one of those have to open up another one of those have to open up another one of those accounts soon. I also don't think that accounts soon. I also don't think that accounts soon. I also don't think that my two hundred thousand dollars of my two hundred thousand dollars of my two hundred thousand dollars of inference is a realistic use case. I inference is a realistic use case. I inference is a realistic use case. I don't think anyone should be doing this don't think anyone should be doing this don't think anyone should be doing this every month. That was more me pushing every month. That was more me pushing every month. That was more me pushing the limits, experimenting, and seeing the limits, experimenting, and seeing the limits, experimenting, and seeing what I could get away with to an extent what I could get away with to an extent what I could get away with to an extent because they've been very tolerant of my because they've been very tolerant of my because they've been very tolerant of my OpenAI has also been too OpenAI has also been too OpenAI has also been too tolerant to the point where others were tolerant to the point where others were tolerant to the point where others were leaking the fact that they had early leaking the fact that they had early leaking the fact that they had early access without permission from OpenAI, access without permission from OpenAI, access without permission from OpenAI, and they didn't do anything about it.

  5. and they didn't do anything about it. and they didn't do anything about it. So, I could have shared my thoughts So, I could have shared my thoughts So, I could have shared my thoughts earlier, but instead I chose to do what earlier, but instead I chose to do what earlier, but instead I chose to do what I thought I should do, which is wait I thought I should do, which is wait I thought I should do, which is wait until the model's officially out for until the model's officially out for until the model's officially out for everyone to access. So again, they're everyone to access. So again, they're everyone to access. So again, they're not seeing this. In fact, there's a good not seeing this. In fact, there's a good not seeing this. In fact, there's a good chance you're watching this before any chance you're watching this before any chance you're watching this before any OpenAI employee has because I just want OpenAI employee has because I just want OpenAI employee has because I just want to share my honest thoughts and to share my honest thoughts and to share my honest thoughts and experience using this model. I'm also experience using this model. I'm also experience using this model. I'm also going to do my best to not talk about going to do my best to not talk about going to do my best to not talk about Fable too much in this video. I'm going Fable too much in this video. I'm going Fable too much in this video. I'm going to do a dedicated comparison later, so to do a dedicated comparison later, so to do a dedicated comparison later, so pardon me for ignoring the existence of pardon me for ignoring the existence of pardon me for ignoring the existence of Fable for a bit. If you do want to hear Fable for a bit. If you do want to hear Fable for a bit. If you do want to hear all of these other things like my proper all of these other things like my proper all of these other things like my proper 5.6 review comparison with Fable, 5.6 review comparison with Fable, 5.6 review comparison with Fable, day-to-day usage tips, trying to day-to-day usage tips, trying to day-to-day usage tips, trying to maximize your utilization, all of the maximize your utilization, all of the maximize your utilization, all of the things that make you really good at things that make you really good at things that make you really good at using this new model without having to using this new model without having to using this new model without having to spend the insane money that I spent on spend the insane money that I spent on spend the insane money that I spent on it, make sure you're subscribed and that it, make sure you're subscribed and that it, make sure you're subscribed and that you hit that bell because I have a lot you hit that bell because I have a lot you hit that bell because I have a lot of videos coming out. I have so many of videos coming out. I have so many of videos coming out. I have so many things I've wanted to talk about with things I've wanted to talk about with things I've wanted to talk about with this model that I haven't been able to, this model that I haven't been able to, this model that I haven't been able to, and I'm finally unleashed. There's going and I'm finally unleashed. There's going and I'm finally unleashed. There's going to be a lot of content the next few to be a lot of content the next few to be a lot of content the next few days, and you're not going to want to days, and you're not going to want to days, and you're not going to want to miss it. I also didn't test the model in miss it. I also didn't test the model in miss it. I also didn't test the model in ChatGPT or 5.6 Pro at all. So, if you're ChatGPT or 5.6 Pro at all. So, if you're ChatGPT or 5.6 Pro at all. So, if you're curious about those, I'll talk about it curious about those, I'll talk about it curious about those, I'll talk about it in the future. As I was saying, so let's in the future. As I was saying, so let's in the future. As I was saying, so let's go through all of the things I did with go through all of the things I did with go through all of the things I did with this model. I worked across 67, nice, this model. I worked across 67, nice, this model. I worked across 67, nice, projects, and I did a lot. Some of them projects, and I did a lot. Some of them projects, and I did a lot. Some of them I did not too many prompts in. I just I did not too many prompts in. I just I did not too many prompts in. I just set off a goal and hoped for the best.

  6. set off a goal and hoped for the best. set off a goal and hoped for the best. Others I put a lot of back and forth Others I put a lot of back and forth Others I put a lot of back and forth effort into. If you were confused why I effort into. If you were confused why I effort into. If you were confused why I had so many different machines on my had so many different machines on my had so many different machines on my network recently, this is a big part of network recently, this is a big part of network recently, this is a big part of why. I wanted to push this model to its why. I wanted to push this model to its why. I wanted to push this model to its limits and really take advantage of the limits and really take advantage of the limits and really take advantage of the generous near unlimited usage they were generous near unlimited usage they were generous near unlimited usage they were giving me. And I did. I pushed it giving me. And I did. I pushed it giving me. And I did. I pushed it really, really hard. It's also worth really, really hard. It's also worth really, really hard. It's also worth noting that I set up my fleet using 5.6, noting that I set up my fleet using 5.6, noting that I set up my fleet using 5.6, their SSH computer use and a centralized their SSH computer use and a centralized their SSH computer use and a centralized config that I had it fan out to all my config that I had it fan out to all my config that I had it fan out to all my machines. It was actually really nice to machines. It was actually really nice to machines. It was actually really nice to work with. And that's one of my favorite work with. And that's one of my favorite work with. And that's one of my favorite things about this model. Not only is it things about this model. Not only is it things about this model. Not only is it much better at understanding intent than much better at understanding intent than much better at understanding intent than 5.5 was, it also just grabs onto your 5.5 was, it also just grabs onto your 5.5 was, it also just grabs onto your task and doesn't let go until it's task and doesn't let go until it's task and doesn't let go until it's finished. This was one of my biggest finished. This was one of my biggest finished. This was one of my biggest issues with 5.5. It almost felt like it issues with 5.5. It almost felt like it issues with 5.5. It almost felt like it would get lost and then stop. Or if would get lost and then stop. Or if would get lost and then stop. Or if anything bad got into its context, you'd anything bad got into its context, you'd anything bad got into its context, you'd have to go reset and make a new thread. have to go reset and make a new thread. have to go reset and make a new thread. 5.6 I did not have that experience with 5.6 I did not have that experience with 5.6 I did not have that experience with it all. I just will let one thread rip it all. I just will let one thread rip it all. I just will let one thread rip forever and it's fine. Even something forever and it's fine. Even something forever and it's fine. Even something like {slash}goal went from holy like {slash}goal went from holy like {slash}goal went from holy disinteresting to me to actually pretty disinteresting to me to actually pretty disinteresting to me to actually pretty exciting cuz the model can manage that exciting cuz the model can manage that exciting cuz the model can manage that and can go for a lot longer without and can go for a lot longer without and can go for a lot longer without issues. That said, I didn't find myself issues. That said, I didn't find myself issues. That said, I didn't find myself reaching for {slash}goal too often reaching for {slash}goal too often reaching for {slash}goal too often because if the model needed to run for because if the model needed to run for because if the model needed to run for more than a few hours, it just could. I more than a few hours, it just could. I more than a few hours, it just could. I would tell it to go do a long-running would tell it to go do a long-running would tell it to go do a long-running task and it will hold on and go for task and it will hold on and go for task and it will hold on and go for hours upon hours. I've had tasks go for hours upon hours. I've had tasks go for hours upon hours. I've had tasks go for 20 plus hours without setting a 20 plus hours without setting a 20 plus hours without setting a {slash}goal at all. I just told it to do {slash}goal at all. I just told it to do {slash}goal at all. I just told it to do a big thing and it went and it did the a big thing and it went and it did the a big thing and it went and it did the big thing. I guess it's time to go over big thing. I guess it's time to go over big thing. I guess it's time to go over all the work I did then, right? I'll all the work I did then, right? I'll all the work I did then, right? I'll start with Lakebed cuz, I'll be frank, I start with Lakebed cuz, I'll be frank, I start with Lakebed cuz, I'll be frank, I really want to get this project out.

  7. really want to get this project out. really want to get this project out. Apparently, if I was paying fast mode Apparently, if I was paying fast mode Apparently, if I was paying fast mode prices, the work I did would have been prices, the work I did would have been prices, the work I did would have been about $30,000 in tokens, which sounds about $30,000 in tokens, which sounds about $30,000 in tokens, which sounds insane. But remember, you don't have to insane. But remember, you don't have to insane. But remember, you don't have to use it in fast mode, which cuts the use it in fast mode, which cuts the use it in fast mode, which cuts the price by almost 2x. And you already get price by almost 2x. And you already get price by almost 2x. And you already get around 14 grand of inference right now around 14 grand of inference right now around 14 grand of inference right now on the $200 a month plan outside of on the $200 a month plan outside of on the $200 a month plan outside of resets using Codex. So like, my usage resets using Codex. So like, my usage resets using Codex. So like, my usage here is actually kind of contained here is actually kind of contained here is actually kind of contained within a normal sub. And that's kind of within a normal sub. And that's kind of within a normal sub. And that's kind of crazy cuz I set up 5.6 on some crazy cuz I set up 5.6 on some crazy cuz I set up 5.6 on some impossible tasks and it went way further impossible tasks and it went way further impossible tasks and it went way further than I ever would have expected. One of than I ever would have expected. One of than I ever would have expected. One of the biggest things I did was move from the biggest things I did was move from the biggest things I did was move from giant monolithic JavaScript files that giant monolithic JavaScript files that giant monolithic JavaScript files that 55 wrote into a properly broken down and 55 wrote into a properly broken down and 55 wrote into a properly broken down and managed TypeScript project. This was a managed TypeScript project. This was a managed TypeScript project. This was a huge change that required a lot of huge change that required a lot of huge change that required a lot of verification and just hands-on work, and verification and just hands-on work, and verification and just hands-on work, and it did all of it without much issue. It it did all of it without much issue. It it did all of it without much issue. It also helped me build a whole new CI also helped me build a whole new CI also helped me build a whole new CI pipeline and preview and release pipeline and preview and release pipeline and preview and release workflows that I'm still using to this workflows that I'm still using to this workflows that I'm still using to this day. I'll probably use indefinitely. It day. I'll probably use indefinitely. It day. I'll probably use indefinitely. It created artifact storage, railway created artifact storage, railway created artifact storage, railway buckets for actually managing files that buckets for actually managing files that buckets for actually managing files that users can upload, fixed a bunch of core users can upload, fixed a bunch of core users can upload, fixed a bunch of core stuff, and also handled a bunch of chaos stuff, and also handled a bunch of chaos stuff, and also handled a bunch of chaos with refreshing deployments with with refreshing deployments with with refreshing deployments with Postgres events. I built CLI login. This Postgres events. I built CLI login. This Postgres events. I built CLI login. This was way more complex than it seems to was way more complex than it seems to was way more complex than it seems to imply here. Building a relationship imply here. Building a relationship imply here. Building a relationship between my production server and the CLI between my production server and the CLI between my production server and the CLI that makes sense that allows you to that makes sense that allows you to that makes sense that allows you to authenticate. Not trivial, and it did authenticate. Not trivial, and it did authenticate. Not trivial, and it did pretty much all of that by itself with pretty much all of that by itself with pretty much all of that by itself with just a little bit of guidance on what I just a little bit of guidance on what I just a little bit of guidance on what I was expecting UX-wise. It implemented was expecting UX-wise. It implemented was expecting UX-wise. It implemented whitelist access, which is a ton of whitelist access, which is a ton of whitelist access, which is a ton of work. Probably the single biggest thing work. Probably the single biggest thing work. Probably the single biggest thing I had it do in Lakebed because it had to I had it do in Lakebed because it had to I had it do in Lakebed because it had to centralize off into its own system to do centralize off into its own system to do centralize off into its own system to do that. I had it fork Shrew my off service that. I had it fork Shrew my off service that. I had it fork Shrew my off service and build that all in, and it works and build that all in, and it works and build that all in, and it works great now. I cannot wait to launch the

  8. great now. I cannot wait to launch the great now. I cannot wait to launch the current staging version of Lakebed as current staging version of Lakebed as current staging version of Lakebed as the production version cuz the vast the production version cuz the vast the production version cuz the vast majority of the changes were written by majority of the changes were written by majority of the changes were written by 56. I hardened the capsule database 56. I hardened the capsule database 56. I hardened the capsule database layer to make the reactivity way more layer to make the reactivity way more layer to make the reactivity way more efficient and reliable. I designed and efficient and reliable. I designed and efficient and reliable. I designed and repeatedly implemented and reviewed the repeatedly implemented and reviewed the repeatedly implemented and reviewed the developer files contract, opaque developer files contract, opaque developer files contract, opaque identifiers, mine policy, all all the identifiers, mine policy, all all the identifiers, mine policy, all all the file object storage stuff cuz I wanted file object storage stuff cuz I wanted file object storage stuff cuz I wanted to make sure that was all really, really to make sure that was all really, really to make sure that was all really, really good and reliable. good and reliable. good and reliable. The first-party off layer, which I The first-party off layer, which I The first-party off layer, which I mentioned before I did for both the mentioned before I did for both the mentioned before I did for both the whitelisting and for the CLI. I had it whitelisting and for the CLI. I had it whitelisting and for the CLI. I had it research different isolate options for research different isolate options for research different isolate options for the runtime that runs the code that the runtime that runs the code that the runtime that runs the code that people ship on Lakebed cuz remember, people ship on Lakebed cuz remember, people ship on Lakebed cuz remember, it's a cloud platform to build real apps it's a cloud platform to build real apps it's a cloud platform to build real apps on. I wanted to make sure if I was to on. I wanted to make sure if I was to on. I wanted to make sure if I was to build my own isolate layer that we did build my own isolate layer that we did build my own isolate layer that we did it right, and it explored a ton of it right, and it explored a ton of it right, and it explored a ton of options and gave me really good HTML options and gave me really good HTML options and gave me really good HTML plans I could read similar to what we're plans I could read similar to what we're plans I could read similar to what we're reading right now. It helped me model reading right now. It helped me model reading right now. It helped me model pricing and quotas, which I'll talk a pricing and quotas, which I'll talk a pricing and quotas, which I'll talk a lot more about my thoughts on what it lot more about my thoughts on what it lot more about my thoughts on what it did here because spoiler, other models did here because spoiler, other models did here because spoiler, other models did this in a I much, much preferred. It did this in a I much, much preferred. It did this in a I much, much preferred. It recovered a ton of PRs for me. I recovered a ton of PRs for me. I recovered a ton of PRs for me. I showcased a few loops I did in the past showcased a few loops I did in the past showcased a few loops I did in the past and to show how powerful they were when and to show how powerful they were when and to show how powerful they were when I had the model spinning up threads to I had the model spinning up threads to I had the model spinning up threads to review PRs and whatnot. When I actually review PRs and whatnot. When I actually review PRs and whatnot. When I actually did that, all of it was 5.6. I had to go did that, all of it was 5.6. I had to go did that, all of it was 5.6. I had to go through and manually swap them all to through and manually swap them all to through and manually swap them all to look like I was using 5.5. I'm sorry I look like I was using 5.5. I'm sorry I look like I was using 5.5. I'm sorry I did that. I did not realize just how did that. I did not realize just how did that. I did not realize just how much worse 5.5 was at sub-agent stuff much worse 5.5 was at sub-agent stuff much worse 5.5 was at sub-agent stuff than 5.6. 5.6 blasted through that. I than 5.6. 5.6 blasted through that. I than 5.6. 5.6 blasted through that. I tried the same prompts later on with 5.5 tried the same prompts later on with 5.5 tried the same prompts later on with 5.5 and it just got confused and lost. It and it just got confused and lost. It and it just got confused and lost. It also did a bunch of auditing for my also did a bunch of auditing for my also did a bunch of auditing for my open-source launch for Lakebed, which open-source launch for Lakebed, which open-source launch for Lakebed, which I'm very, very excited about. It landed I'm very, very excited about. It landed I'm very, very excited about. It landed a ton of code, dozens of PRs, and really a ton of code, dozens of PRs, and really a ton of code, dozens of PRs, and really helped to turn Lakebed from a weird helped to turn Lakebed from a weird helped to turn Lakebed from a weird small side project into a thing I'm

  9. small side project into a thing I'm small side project into a thing I'm excited to ship. And I'm so excited to excited to ship. And I'm so excited to excited to ship. And I'm so excited to ship it. You guys have no idea. Then, of ship it. You guys have no idea. Then, of ship it. You guys have no idea. Then, of course, we have T3 Code, the thing that course, we have T3 Code, the thing that course, we have T3 Code, the thing that I did a lot of my work with. T3 Code is I did a lot of my work with. T3 Code is I did a lot of my work with. T3 Code is the project Julius and I have been the project Julius and I have been the project Julius and I have been working on non-stop to make it easier to working on non-stop to make it easier to working on non-stop to make it easier to manage your agents on whatever machine manage your agents on whatever machine manage your agents on whatever machine they're on from a unified layer, whether they're on from a unified layer, whether they're on from a unified layer, whether that is the desktop app, the web app, or that is the desktop app, the web app, or that is the desktop app, the web app, or a phone app, which is actually very, a phone app, which is actually very, a phone app, which is actually very, very far along, too. Early access for very far along, too. Early access for very far along, too. Early access for the mobile app coming sooner than you the mobile app coming sooner than you the mobile app coming sooner than you think. My work on T3 Code showcased just think. My work on T3 Code showcased just think. My work on T3 Code showcased just how thorough and thoughtful 5.6 can be, how thorough and thoughtful 5.6 can be, how thorough and thoughtful 5.6 can be, in particular with mobile. I've had so in particular with mobile. I've had so in particular with mobile. I've had so many problems with different models on many problems with different models on many problems with different models on mobile, even modern, really high-end mobile, even modern, really high-end mobile, even modern, really high-end ones that I'm trying to not just ones that I'm trying to not just ones that I'm trying to not just name-drop constantly. 5.6 is so much name-drop constantly. 5.6 is so much name-drop constantly. 5.6 is so much better at mobile that I realized I better at mobile that I realized I better at mobile that I realized I needed to push it harder, so I did. I needed to push it harder, so I did. I needed to push it harder, so I did. I had it rebuild the entire React Native had it rebuild the entire React Native had it rebuild the entire React Native app using AppKit and Swift, as well as app using AppKit and Swift, as well as app using AppKit and Swift, as well as SwiftUI separately. So, that's two SwiftUI separately. So, that's two SwiftUI separately. So, that's two native rewrites from scratch. And I was native rewrites from scratch. And I was native rewrites from scratch. And I was blown away with both. They were actually blown away with both. They were actually blown away with both. They were actually fully complete, end-to-end, functioning fully complete, end-to-end, functioning fully complete, end-to-end, functioning with all the same features that existed with all the same features that existed with all the same features that existed in the React Native version. And it did in the React Native version. And it did in the React Native version. And it did that in like two to four hours for each that in like two to four hours for each that in like two to four hours for each of them. Unbelievable. I thought it of them. Unbelievable. I thought it of them. Unbelievable. I thought it would take a day and make something would take a day and make something would take a day and make something entirely broken. But when you combine entirely broken. But when you combine entirely broken. But when you combine the capability it has with mobile with the capability it has with mobile with the capability it has with mobile with its computer use capability, where I can its computer use capability, where I can its computer use capability, where I can actually spin up the simulator and check actually spin up the simulator and check actually spin up the simulator and check things in it, you can get full things in it, you can get full things in it, you can get full end-to-end insane projects like this end-to-end insane projects like this end-to-end insane projects like this functioning. I had a bunch of other work functioning. I had a bunch of other work functioning. I had a bunch of other work I did with 5-6 in T3 code, but I really I did with 5-6 in T3 code, but I really I did with 5-6 in T3 code, but I really want to lean into this rewrite angle want to lean into this rewrite angle want to lean into this rewrite angle because I was blown away with how much I because I was blown away with how much I because I was blown away with how much I could rebuild from scratch using the could rebuild from scratch using the could rebuild from scratch using the model. I've a lot more to say about

  10. model. I've a lot more to say about model. I've a lot more to say about these types of bold rewrites and believe these types of bold rewrites and believe these types of bold rewrites and believe me, we'll be back to that. First, I want me, we'll be back to that. First, I want me, we'll be back to that. First, I want to go over the other things I built with to go over the other things I built with to go over the other things I built with T3 code in 5-6. I built a full T3 code in 5-6. I built a full T3 code in 5-6. I built a full end-to-end computer use loop into T3 end-to-end computer use loop into T3 end-to-end computer use loop into T3 code so that you could do this type of code so that you could do this type of code so that you could do this type of iteration with mobile and computer use iteration with mobile and computer use iteration with mobile and computer use with 5-6 in T3 code directly. Ended up with 5-6 in T3 code directly. Ended up with 5-6 in T3 code directly. Ended up not even putting up the PR because not even putting up the PR because not even putting up the PR because Julius got it working even better Julius got it working even better Julius got it working even better himself, again using the same model. We himself, again using the same model. We himself, again using the same model. We have audited a collapse worked for have audited a collapse worked for have audited a collapse worked for history, which was very, very nice history, which was very, very nice history, which was very, very nice because we had too much stuff going on because we had too much stuff going on because we had too much stuff going on in the history. Audited the Codex app in the history. Audited the Codex app in the history. Audited the Codex app server primitives in order to try and server primitives in order to try and server primitives in order to try and make sub-agents work better. I even got make sub-agents work better. I even got make sub-agents work better. I even got a PR up that did that. I don't know what a PR up that did that. I don't know what a PR up that did that. I don't know what we'll do with it. Designed a route back we'll do with it. Designed a route back we'll do with it. Designed a route back top bar computer switcher as well cuz I top bar computer switcher as well cuz I top bar computer switcher as well cuz I wanted to try and manage all my wanted to try and manage all my wanted to try and manage all my different computers in T3 code different computers in T3 code different computers in T3 code differently. Didn't end up shipping differently. Didn't end up shipping differently. Didn't end up shipping this, but it was insane how far it got this, but it was insane how far it got this, but it was insane how far it got without anything beyond a vague prompt without anything beyond a vague prompt without anything beyond a vague prompt of what I was thinking of. Code thing is of what I was thinking of. Code thing is of what I was thinking of. Code thing is also T3 code. It just didn't split also T3 code. It just didn't split also T3 code. It just didn't split correctly when I asked it to combine all correctly when I asked it to combine all correctly when I asked it to combine all these projects. I rebuilt the marketing these projects. I rebuilt the marketing these projects. I rebuilt the marketing site and got it to the stage it's in site and got it to the stage it's in site and got it to the stage it's in today using the model. So, if you're today using the model. So, if you're today using the model. So, if you're worried about how good this model is at worried about how good this model is at worried about how good this model is at front end, I'm not going to say it's front end, I'm not going to say it's front end, I'm not going to say it's perfect, but when it's steered the right perfect, but when it's steered the right perfect, but when it's steered the right way, you can get good front ends out of way, you can get good front ends out of way, you can get good front ends out of it. I think this looks pretty good and it. I think this looks pretty good and it. I think this looks pretty good and most of this was done with 5-6.

  11. most of this was done with 5-6. most of this was done with 5-6. So, yeah, take that as you will. I also So, yeah, take that as you will. I also So, yeah, take that as you will. I also had it use sub-agents to go through all had it use sub-agents to go through all had it use sub-agents to go through all of the T3 code PRs and help me of the T3 code PRs and help me of the T3 code PRs and help me orchestrate all of them to figure out orchestrate all of them to figure out orchestrate all of them to figure out what to prioritize and merge and not what to prioritize and merge and not what to prioritize and merge and not merge. I ended up liking the results merge. I ended up liking the results merge. I ended up liking the results here so much that I built an automation here so much that I built an automation here so much that I built an automation with Hermes to do this every single day. with Hermes to do this every single day. with Hermes to do this every single day. It has helped me stay on top of PRs It has helped me stay on top of PRs It has helped me stay on top of PRs better than I ever have in my life. Here better than I ever have in my life. Here better than I ever have in my life. Here is the mobile native version. As I is the mobile native version. As I is the mobile native version. As I mentioned before, I rebuilt the whole mentioned before, I rebuilt the whole mentioned before, I rebuilt the whole thing in Swift UI and it succeeded. Here thing in Swift UI and it succeeded. Here thing in Swift UI and it succeeded. Here is it talking about that. I also just is it talking about that. I also just is it talking about that. I also just had it do a bunch of profiling on the had it do a bunch of profiling on the had it do a bunch of profiling on the web app to find any performance issues web app to find any performance issues web app to find any performance issues that are worth addressing. Okay, I need that are worth addressing. Okay, I need that are worth addressing. Okay, I need to talk about the rewrite thing because to talk about the rewrite thing because to talk about the rewrite thing because it is just so cool. The combination of it is just so cool. The combination of it is just so cool. The combination of the model working well for these types the model working well for these types the model working well for these types of rewrites and the inspiration I had of rewrites and the inspiration I had of rewrites and the inspiration I had from a certain bun being rewritten in a from a certain bun being rewritten in a from a certain bun being rewritten in a certain rust with a certain fable had me certain rust with a certain fable had me certain rust with a certain fable had me inspired to try doing similar rewrites. inspired to try doing similar rewrites. inspired to try doing similar rewrites. So I did. So I did. So I did. I started by trying to rewrite Hermes I started by trying to rewrite Hermes I started by trying to rewrite Hermes agent in rust with complete parody and agent in rust with complete parody and agent in rust with complete parody and ended up getting pretty damn far. One of ended up getting pretty damn far. One of ended up getting pretty damn far. One of the interesting things I did when I was the interesting things I did when I was the interesting things I did when I was working on this is I had it archive all working on this is I had it archive all working on this is I had it archive all of the stuff I had done with my Hermes of the stuff I had done with my Hermes of the stuff I had done with my Hermes agent as well as Ben with his. So I had agent as well as Ben with his. So I had agent as well as Ben with his. So I had all of our history and then I had 56 go all of our history and then I had 56 go all of our history and then I had 56 go through it and figure out all of the through it and figure out all of the through it and figure out all of the features that we need in Hermes in order features that we need in Hermes in order features that we need in Hermes in order to use a port of it and it came up with to use a port of it and it came up with to use a port of it and it came up with the 50 or so percent of features that we the 50 or so percent of features that we the 50 or so percent of features that we actually used and went and implemented actually used and went and implemented actually used and went and implemented almost all of them. I actually got it almost all of them. I actually got it almost all of them. I actually got it spun up in its own separate channel and spun up in its own separate channel and spun up in its own separate channel and responding creating threads using all of responding creating threads using all of responding creating threads using all of its skills calling the model and doing its skills calling the model and doing its skills calling the model and doing the things I expected to do with a goal the things I expected to do with a goal the things I expected to do with a goal I ran for about a day. It works. A full

  12. I ran for about a day. It works. A full I ran for about a day. It works. A full from the ground up rewrite in rust that from the ground up rewrite in rust that from the ground up rewrite in rust that uses like 15 megabytes of RAM instead of uses like 15 megabytes of RAM instead of uses like 15 megabytes of RAM instead of the giant Python version that is not the the giant Python version that is not the the giant Python version that is not the most performant thing but that's not the most performant thing but that's not the most performant thing but that's not the goal of Hermes. I have it on a dedicated goal of Hermes. I have it on a dedicated goal of Hermes. I have it on a dedicated Mac Mini anyways, who cares. I was just Mac Mini anyways, who cares. I was just Mac Mini anyways, who cares. I was just curious what it would look like to do curious what it would look like to do curious what it would look like to do this and make a very small memory safe this and make a very small memory safe this and make a very small memory safe reliable minimal Hermes agent style reliable minimal Hermes agent style reliable minimal Hermes agent style thing that I could put in a small VM and thing that I could put in a small VM and thing that I could put in a small VM and throw wherever and it went way better throw wherever and it went way better throw wherever and it went way better than I expected. If I was to really sit than I expected. If I was to really sit than I expected. If I was to really sit there and work on it further, I bet I there and work on it further, I bet I there and work on it further, I bet I could finish this without too much could finish this without too much could finish this without too much additional time. Not anywhere near additional time. Not anywhere near additional time. Not anywhere near feature complete against Hermes agent feature complete against Hermes agent feature complete against Hermes agent but to a point where I would be able to but to a point where I would be able to but to a point where I would be able to use it every day without issue. To be use it every day without issue. To be use it every day without issue. To be fair, that would have been about 13 fair, that would have been about 13 fair, that would have been about 13 grand on fast mode probably closer to grand on fast mode probably closer to grand on fast mode probably closer to five to seven if it wasn't. Still pretty five to seven if it wasn't. Still pretty five to seven if it wasn't. Still pretty insane. Not as insane as this port insane. Not as insane as this port insane. Not as insane as this port though and this has been very hard for though and this has been very hard for though and this has been very hard for me to not talk about because I had this me to not talk about because I had this me to not talk about because I had this one going for multiple days. I started one going for multiple days. I started one going for multiple days. I started rebuilding the TypeScript Go port in rebuilding the TypeScript Go port in rebuilding the TypeScript Go port in rust. Not because I thought it would rust. Not because I thought it would rust. Not because I thought it would work or I thought it would be useful, work or I thought it would be useful, work or I thought it would be useful, just cuz I was curious how it would go. just cuz I was curious how it would go. just cuz I was curious how it would go. I was actually able to get a 100% I was actually able to get a 100% I was actually able to get a 100% working TypeScript compiler, working TypeScript compiler, working TypeScript compiler, specifically the transpiler that turns specifically the transpiler that turns specifically the transpiler that turns the TypeScript into JavaScript, all the TypeScript into JavaScript, all the TypeScript into JavaScript, all written in Rust that is up to 18 times written in Rust that is up to 18 times written in Rust that is up to 18 times faster than the equivalent Go version.

  13. faster than the equivalent Go version. faster than the equivalent Go version. But, I didn't get the type checking very But, I didn't get the type checking very But, I didn't get the type checking very far. I tried and I had a lot of tests far. I tried and I had a lot of tests far. I tried and I had a lot of tests passing, more than I would have passing, more than I would have passing, more than I would have expected. The code base that it built is expected. The code base that it built is expected. The code base that it built is pretty wild. It ended up being almost pretty wild. It ended up being almost pretty wild. It ended up being almost 200,000 lines of Rust and a lot of tests 200,000 lines of Rust and a lot of tests 200,000 lines of Rust and a lot of tests passing, although admittedly, according passing, although admittedly, according passing, although admittedly, according to Fable, which I just had analyzed the to Fable, which I just had analyzed the to Fable, which I just had analyzed the code, it's a broad prototype with a tiny code, it's a broad prototype with a tiny code, it's a broad prototype with a tiny verified slice and it's far from verified slice and it's far from verified slice and it's far from shippable. It's 15 to 20% of the way to shippable. It's 15 to 20% of the way to shippable. It's 15 to 20% of the way to a usable tool, but only 5% of the way to a usable tool, but only 5% of the way to a usable tool, but only 5% of the way to a proper TS Go replacement despite over a proper TS Go replacement despite over a proper TS Go replacement despite over 195,000 lines of Rust across the 29 195,000 lines of Rust across the 29 195,000 lines of Rust across the 29 crates that it built. It does have a crates that it built. It does have a crates that it built. It does have a genuinely working deliberately narrow no genuinely working deliberately narrow no genuinely working deliberately narrow no check transpiler from a five file check transpiler from a five file check transpiler from a five file corpus, but everything else is code that corpus, but everything else is code that corpus, but everything else is code that hasn't been proven and the tests are hasn't been proven and the tests are hasn't been proven and the tests are currently failing. Currently, I have a currently failing. Currently, I have a currently failing. Currently, I have a bunch of work that was stuck in a stash, bunch of work that was stuck in a stash, bunch of work that was stuck in a stash, so I'm going to clean that up and figure so I'm going to clean that up and figure so I'm going to clean that up and figure out how far it actually is of the out how far it actually is of the out how far it actually is of the Obviously, I haven't had time to Obviously, I haven't had time to Obviously, I haven't had time to actually play with this, so we'll see actually play with this, so we'll see actually play with this, so we'll see where that ends up. I'll show more in a where that ends up. I'll show more in a where that ends up. I'll show more in a minute, but I want to talk about the minute, but I want to talk about the minute, but I want to talk about the other things I worked on first. If you other things I worked on first. If you other things I worked on first. If you saw my video where I dropped all the saw my video where I dropped all the saw my video where I dropped all the ideas I wish someone would build, you ideas I wish someone would build, you ideas I wish someone would build, you might remember the idea of a might remember the idea of a might remember the idea of a Dropbox-like cloud that lets you sync Dropbox-like cloud that lets you sync Dropbox-like cloud that lets you sync your dev folder across machines. You your dev folder across machines. You your dev folder across machines. You probably understand why I wanted this probably understand why I wanted this probably understand why I wanted this now cuz I had all this work across all now cuz I had all this work across all now cuz I had all this work across all these computers, but I also decided to these computers, but I also decided to these computers, but I also decided to just throw out a goal and see what just throw out a goal and see what just throw out a goal and see what happens. Before this, I hadn't actually happens. Before this, I hadn't actually happens. Before this, I hadn't actually analyzed how many tokens that one goal analyzed how many tokens that one goal analyzed how many tokens that one goal run burned. Turned out it was 71.2 run burned. Turned out it was 71.2 run burned. Turned out it was 71.2 billion, which at the fast mode pricing billion, which at the fast mode pricing billion, which at the fast mode pricing would have been around $91,000.

  14. would have been around $91,000. would have been around $91,000. I had not went to go check and see how I had not went to go check and see how I had not went to go check and see how well this works. I haven't even opened well this works. I haven't even opened well this works. I haven't even opened the project. I just let the goal run the project. I just let the goal run the project. I just let the goal run nearly forever, and I guess it did. I nearly forever, and I guess it did. I nearly forever, and I guess it did. I actually learned that it was using actually learned that it was using actually learned that it was using PlanetScale when one of my friends who PlanetScale when one of my friends who PlanetScale when one of my friends who worked there mentioned, "Yo, I heard you worked there mentioned, "Yo, I heard you worked there mentioned, "Yo, I heard you were working on FS2 again. We saw you were working on FS2 again. We saw you were working on FS2 again. We saw you registered on PlanetScale. How has it registered on PlanetScale. How has it registered on PlanetScale. How has it been?" I was like, "Wait, I did?" And been?" I was like, "Wait, I did?" And been?" I was like, "Wait, I did?" And then I went and checked the logs and I then I went and checked the logs and I then I went and checked the logs and I saw it had autonomously registered for saw it had autonomously registered for saw it had autonomously registered for PlanetScale itself in order to build PlanetScale itself in order to build PlanetScale itself in order to build what it was trying to build. I'm what it was trying to build. I'm what it was trying to build. I'm actually much more excited to go try actually much more excited to go try actually much more excited to go try this out now cuz I did not think it this out now cuz I did not think it this out now cuz I did not think it would get as far as it did. So, uh would get as far as it did. So, uh would get as far as it did. So, uh fingers crossed it went well. I'll be fingers crossed it went well. I'll be fingers crossed it went well. I'll be sure to check it in the near future. I sure to check it in the near future. I sure to check it in the near future. I have to get it actually deployed though, have to get it actually deployed though, have to get it actually deployed though, which will be a lot easier because the which will be a lot easier because the which will be a lot easier because the computer use is phenomenal. So, I'm computer use is phenomenal. So, I'm computer use is phenomenal. So, I'm really, really excited to see if it can really, really excited to see if it can really, really excited to see if it can not just set up this project for me, but not just set up this project for me, but not just set up this project for me, but actually go into all the dashboards and actually go into all the dashboards and actually go into all the dashboards and set up all of the things it needs to set up all of the things it needs to set up all of the things it needs to really deploy it. Ooh, the bootable really deploy it. Ooh, the bootable really deploy it. Ooh, the bootable Codex drive was actually one of my Codex drive was actually one of my Codex drive was actually one of my favorite things I did. Since I was favorite things I did. Since I was favorite things I did. Since I was managing all these machines on my managing all these machines on my managing all these machines on my network, I needed to make sure I could network, I needed to make sure I could network, I needed to make sure I could actually set them up properly. And I was actually set them up properly. And I was actually set them up properly. And I was getting tired of reconfiguring whatever getting tired of reconfiguring whatever getting tired of reconfiguring whatever Linux distro over and over again, and Linux distro over and over again, and Linux distro over and over again, and also having to recover it when they also having to recover it when they also having to recover it when they broke. When I was ripping an SSD out of broke. When I was ripping an SSD out of broke. When I was ripping an SSD out of one of the machines, it didn't properly one of the machines, it didn't properly one of the machines, it didn't properly reconfigure the boot partitions on the reconfigure the boot partitions on the reconfigure the boot partitions on the existing SSD that I left in the machine.

  15. existing SSD that I left in the machine. existing SSD that I left in the machine. All of the content it needed was there, All of the content it needed was there, All of the content it needed was there, but it wasn't actually pointing at the but it wasn't actually pointing at the but it wasn't actually pointing at the right partitions, so it open up an empty right partitions, so it open up an empty right partitions, so it open up an empty grub that could only run memory tests. I grub that could only run memory tests. I grub that could only run memory tests. I realized I wanted an easy bootable drive realized I wanted an easy bootable drive realized I wanted an easy bootable drive that I could use with Codex and Claude that I could use with Codex and Claude that I could use with Codex and Claude already off, so I could plug in, boot to already off, so I could plug in, boot to already off, so I could plug in, boot to that, and then use it to fix things. But that, and then use it to fix things. But that, and then use it to fix things. But setting up that image was taking a setting up that image was taking a setting up that image was taking a while. So, while I had one instance of while. So, while I had one instance of while. So, while I had one instance of Codex configuring this flash drive for Codex configuring this flash drive for Codex configuring this flash drive for what I wanted here, which it did, and what I wanted here, which it did, and what I wanted here, which it did, and it's awesome by the way. I highly it's awesome by the way. I highly it's awesome by the way. I highly recommend setting up a drive like this, recommend setting up a drive like this, recommend setting up a drive like this, where you can boot any machine to it. where you can boot any machine to it. where you can boot any machine to it. Not a Mac, of course, cuz Macs don't Not a Mac, of course, cuz Macs don't Not a Mac, of course, cuz Macs don't boot to Linux nowadays particularly boot to Linux nowadays particularly boot to Linux nowadays particularly well. But all the other machines on my well. But all the other machines on my well. But all the other machines on my network, I could plug in the drive, network, I could plug in the drive, network, I could plug in the drive, reboot, boot to the USB, connect the reboot, boot to the USB, connect the reboot, boot to the USB, connect the network, and now I have remote access to network, and now I have remote access to network, and now I have remote access to that machine with Codex and Claude that machine with Codex and Claude that machine with Codex and Claude already set up. Super cool. But I didn't already set up. Super cool. But I didn't already set up. Super cool. But I didn't even end up needing it for the box I was even end up needing it for the box I was even end up needing it for the box I was trying to repair, because once I booted trying to repair, because once I booted trying to repair, because once I booted it, I was asking Codex for help like it, I was asking Codex for help like it, I was asking Codex for help like getting around weird BIOS things cuz getting around weird BIOS things cuz getting around weird BIOS things cuz HP's BIOS is trash. It kept HP's BIOS is trash. It kept HP's BIOS is trash. It kept hallucinating things I could do in the hallucinating things I could do in the hallucinating things I could do in the BIOS. So, eventually I said, "Okay, BIOS. So, eventually I said, "Okay, BIOS. So, eventually I said, "Okay, here, have direct access." I used my here, have direct access." I used my here, have direct access." I used my remote KVM, which lets me connect to the remote KVM, which lets me connect to the remote KVM, which lets me connect to the HDMI and a fake keyboard and mouse, so I HDMI and a fake keyboard and mouse, so I HDMI and a fake keyboard and mouse, so I can access the computer fully through can access the computer fully through can access the computer fully through the web, including the actual BIOS and the web, including the actual BIOS and the web, including the actual BIOS and config, and all of those types of config, and all of those types of config, and all of those types of things. It lets me reboot a machine things. It lets me reboot a machine things. It lets me reboot a machine remotely, super helpful. But, you also remotely, super helpful. But, you also remotely, super helpful. But, you also have to time a bunch of presses in order have to time a bunch of presses in order have to time a bunch of presses in order to do things like get to a specific boot to do things like get to a specific boot to do things like get to a specific boot menu or get around grub. So, I just menu or get around grub. So, I just menu or get around grub. So, I just figured I would have to do those parts, figured I would have to do those parts, figured I would have to do those parts, and maybe Codex would be able to help.

  16. and maybe Codex would be able to help. and maybe Codex would be able to help. When 5 6 started hallucinating those When 5 6 started hallucinating those When 5 6 started hallucinating those weird BIOS things that didn't actually weird BIOS things that didn't actually weird BIOS things that didn't actually exist, I told it to use the tab in exist, I told it to use the tab in exist, I told it to use the tab in helium and look and see and click around helium and look and see and click around helium and look and see and click around to see if it could find what it thought to see if it could find what it thought to see if it could find what it thought existed. And it failed. And then it existed. And it failed. And then it existed. And it failed. And then it rebooted, and it got into the broken rebooted, and it got into the broken rebooted, and it got into the broken grub state. So, I told it that it was in grub state. So, I told it that it was in grub state. So, I told it that it was in that broken grub state and asked for that broken grub state and asked for that broken grub state and asked for help fixing it. help fixing it. help fixing it. And it effectively said, "Hold my beer." And it effectively said, "Hold my beer." And it effectively said, "Hold my beer." Rebooted a few times, got into a shell Rebooted a few times, got into a shell Rebooted a few times, got into a shell in grub, booted correctly, and then used in grub, booted correctly, and then used in grub, booted correctly, and then used the computer used to remote control the the computer used to remote control the the computer used to remote control the computer, opened the terminal, and then computer, opened the terminal, and then computer, opened the terminal, and then fixed the boot partitions. All fixed the boot partitions. All fixed the boot partitions. All autonomously. These are the types of autonomously. These are the types of autonomously. These are the types of tasks that used to give me nightmares. I tasks that used to give me nightmares. I tasks that used to give me nightmares. I know how to do them, mostly, but they know how to do them, mostly, but they know how to do them, mostly, but they require a lot of research and patience require a lot of research and patience require a lot of research and patience and risk tolerance. and risk tolerance. and risk tolerance. 5 6 just went out and did it with no 5 6 just went out and did it with no 5 6 just went out and did it with no additional effort. I still struggle to additional effort. I still struggle to additional effort. I still struggle to believe that it was actually capable of believe that it was actually capable of believe that it was actually capable of doing that. It was one of the most like, doing that. It was one of the most like, doing that. It was one of the most like, "Oh, we're really there." moments "Oh, we're really there." moments "Oh, we're really there." moments I've ever had, and it made me start I've ever had, and it made me start I've ever had, and it made me start pushing computer use way harder. pushing computer use way harder. pushing computer use way harder. And it did all of this faster than it And it did all of this faster than it And it did all of this faster than it was able to set up that recovery boot was able to set up that recovery boot was able to set up that recovery boot drive. Both of which were unbelievably drive. Both of which were unbelievably drive. Both of which were unbelievably cool, by the way. I use both a lot. I cool, by the way. I use both a lot. I cool, by the way. I use both a lot. I have computer use through the GL.iNet have computer use through the GL.iNet have computer use through the GL.iNet KVM, which is still one of my favorite KVM, which is still one of my favorite KVM, which is still one of my favorite devices. I'll have a link in the devices. I'll have a link in the devices. I'll have a link in the description if you want to grab one cuz description if you want to grab one cuz description if you want to grab one cuz it's so helpful. But, I also got the it's so helpful. But, I also got the it's so helpful. But, I also got the boot drive, and I can use both of these boot drive, and I can use both of these boot drive, and I can use both of these to configure my network and all the to configure my network and all the to configure my network and all the machines on it. I also have a repo that machines on it. I also have a repo that machines on it. I also have a repo that I made called fleet that organizes all I made called fleet that organizes all I made called fleet that organizes all of the different computers I have, what of the different computers I have, what of the different computers I have, what their purposes are, what's configured on their purposes are, what's configured on their purposes are, what's configured on them, and what my default configs are.

  17. them, and what my default configs are. them, and what my default configs are. So, when I get a new machine, I just So, when I get a new machine, I just So, when I get a new machine, I just open up that repo on this one, I tell it open up that repo on this one, I tell it open up that repo on this one, I tell it what I got and why I want to set it up, what I got and why I want to set it up, what I got and why I want to set it up, and what I want it to do, and then it and what I want it to do, and then it and what I want it to do, and then it uses all of that context to configure uses all of that context to configure uses all of that context to configure that machine exactly how I want without that machine exactly how I want without that machine exactly how I want without me even having to touch it. It has been me even having to touch it. It has been me even having to touch it. It has been so nice. I keep remembering random so nice. I keep remembering random so nice. I keep remembering random things I had the model do that impressed things I had the model do that impressed things I had the model do that impressed me. Like, I had it overhaul skate bench me. Like, I had it overhaul skate bench me. Like, I had it overhaul skate bench again and it did a great job. It's way again and it did a great job. It's way again and it did a great job. It's way more reliable and easy for me to control more reliable and easy for me to control more reliable and easy for me to control now. There is still a bug with the now. There is still a bug with the now. There is still a bug with the reporting for token spend here cuz I reporting for token spend here cuz I reporting for token spend here cuz I know that those numbers are very very know that those numbers are very very know that those numbers are very very wrong, especially when you compare it wrong, especially when you compare it wrong, especially when you compare it with the costs. with the costs. with the costs. Not very good skate bench score to be Not very good skate bench score to be Not very good skate bench score to be fair, but that's not what we're here to fair, but that's not what we're here to fair, but that's not what we're here to talk about. I'm here to talk about how talk about. I'm here to talk about how talk about. I'm here to talk about how useful these are for coding. I had it useful these are for coding. I had it useful these are for coding. I had it build a 3D version of fish slop that I build a 3D version of fish slop that I build a 3D version of fish slop that I currently can't find on my machine, so currently can't find on my machine, so currently can't find on my machine, so I'll have to figure out where that is I'll have to figure out where that is I'll have to figure out where that is later. I am sorry. I was really thinking later. I am sorry. I was really thinking later. I am sorry. I was really thinking I'd be able to grab that, but apparently I'd be able to grab that, but apparently I'd be able to grab that, but apparently when I CD'n it just doesn't exist. Oh, I when I CD'n it just doesn't exist. Oh, I when I CD'n it just doesn't exist. Oh, I had to help a bunch with prime day had to help a bunch with prime day had to help a bunch with prime day stuff. I do a prime day thread cuz I am stuff. I do a prime day thread cuz I am stuff. I do a prime day thread cuz I am a nerd for hunting deals and it was able a nerd for hunting deals and it was able a nerd for hunting deals and it was able to verify a lot of the deals I found, to verify a lot of the deals I found, to verify a lot of the deals I found, but more importantly create genius but more importantly create genius but more importantly create genius links, which is my link tracker that links, which is my link tracker that links, which is my link tracker that makes it easier to redirect to the makes it easier to redirect to the makes it easier to redirect to the country of choice for whatever country country of choice for whatever country country of choice for whatever country you guys are in. That's normally a very you guys are in. That's normally a very you guys are in. That's normally a very slow manual process clicking around the slow manual process clicking around the slow manual process clicking around the web. It automated the whole thing web. It automated the whole thing web. It automated the whole thing through browser use for me and responded through browser use for me and responded through browser use for me and responded with all of my links ready to go. It was with all of my links ready to go. It was with all of my links ready to go. It was so nice. It also wrote some copy for me so nice. It also wrote some copy for me so nice. It also wrote some copy for me that was absolute garbage, so I threw that was absolute garbage, so I threw that was absolute garbage, so I threw all of that out, but the rest was all of that out, but the rest was all of that out, but the rest was impressive. It built my little terminal impressive. It built my little terminal impressive. It built my little terminal token tracker, which I'm sure Phase can token tracker, which I'm sure Phase can token tracker, which I'm sure Phase can sneak a picture of somewhere in here. I sneak a picture of somewhere in here. I sneak a picture of somewhere in here. I showed it off in the podcast, which by showed it off in the podcast, which by showed it off in the podcast, which by the way, if you want to hear me yap the way, if you want to hear me yap the way, if you want to hear me yap about this model even more, I have a about this model even more, I have a about this model even more, I have a whole podcast episode where I do that,

  18. whole podcast episode where I do that, whole podcast episode where I do that, too. Filmed with Ben, who also had early too. Filmed with Ben, who also had early too. Filmed with Ben, who also had early access weeks before the model was access weeks before the model was access weeks before the model was officially announced. Very fun episode. officially announced. Very fun episode. officially announced. Very fun episode. I have learned a lot more about it I have learned a lot more about it I have learned a lot more about it since, but I think that was a good like since, but I think that was a good like since, but I think that was a good like oh podcast episode for us where you oh podcast episode for us where you oh podcast episode for us where you get to see our genuine reactions. But I get to see our genuine reactions. But I get to see our genuine reactions. But I set up that terminal with all of the set up that terminal with all of the set up that terminal with all of the content on it and its ability to get content on it and its ability to get content on it and its ability to get that content from my fleet using this that content from my fleet using this that content from my fleet using this model. It did a very good job, but its model. It did a very good job, but its model. It did a very good job, but its UI was incredibly ugly and I had to UI was incredibly ugly and I had to UI was incredibly ugly and I had to rebuild what it did there with Opus 48. rebuild what it did there with Opus 48. rebuild what it did there with Opus 48. Spoiler, this model isn't magically Spoiler, this model isn't magically Spoiler, this model isn't magically perfect at front end like other models perfect at front end like other models perfect at front end like other models can be. It's much more steerable and has can be. It's much more steerable and has can be. It's much more steerable and has less bad of taste by default, still not less bad of taste by default, still not less bad of taste by default, still not a great front end model by any stretch a great front end model by any stretch a great front end model by any stretch of the imagination. I was wrong. Turns of the imagination. I was wrong. Turns of the imagination. I was wrong. Turns out I do actually have a version of to out I do actually have a version of to out I do actually have a version of to slop here that I haven't seen before. It slop here that I haven't seen before. It slop here that I haven't seen before. It generated all of the art itself, too. generated all of the art itself, too. generated all of the art itself, too. It's not great, but it's passable. UI is It's not great, but it's passable. UI is It's not great, but it's passable. UI is a bit garbage though, if I'm being a bit garbage though, if I'm being a bit garbage though, if I'm being honest. Want to find the 3D version honest. Want to find the 3D version honest. Want to find the 3D version though. Give me a sec to do that. And though. Give me a sec to do that. And though. Give me a sec to do that. And here is the 3D version where it made here is the 3D version where it made here is the 3D version where it made everything itself. It made the 3D everything itself. It made the 3D everything itself. It made the 3D environment. It made all of the rocks on environment. It made all of the rocks on environment. It made all of the rocks on the ground. It made its own textures.

  19. the ground. It made its own textures. the ground. It made its own textures. Some of the textures are better than Some of the textures are better than Some of the textures are better than what I got out of something like Grok, what I got out of something like Grok, what I got out of something like Grok, but not all of them. I think the rocks but not all of them. I think the rocks but not all of them. I think the rocks are actually quite a bit worse. are actually quite a bit worse. are actually quite a bit worse. It does get the controls a lot better It does get the controls a lot better It does get the controls a lot better though. I had to push it hard to make though. I had to push it hard to make though. I had to push it hard to make the controls more like a Subnautica the controls more like a Subnautica the controls more like a Subnautica style like adventure where you're style like adventure where you're style like adventure where you're steering and driving your ship. steering and driving your ship. steering and driving your ship. It also screwed up the shoot controls a It also screwed up the shoot controls a It also screwed up the shoot controls a bunch and it made some of the weirdest bunch and it made some of the weirdest bunch and it made some of the weirdest most hideous models for the monsters and most hideous models for the monsters and most hideous models for the monsters and for the fish. for the fish. for the fish. It It's understanding of 3D is uh the It It's understanding of 3D is uh the It It's understanding of 3D is uh the best I can put it is strange. best I can put it is strange. best I can put it is strange. But uh But uh But uh yeah, this is one of those areas where yeah, this is one of those areas where yeah, this is one of those areas where like the progress is real. It's still like the progress is real. It's still like the progress is real. It's still not at the point where I would want it not at the point where I would want it not at the point where I would want it to ship this at all, but it is to ship this at all, but it is to ship this at all, but it is progressing fast enough that I'm progressing fast enough that I'm progressing fast enough that I'm definitely keeping my eye on these types definitely keeping my eye on these types definitely keeping my eye on these types of like 3D game use cases. I did a bunch of like 3D game use cases. I did a bunch of like 3D game use cases. I did a bunch of other weird things like trying to of other weird things like trying to of other weird things like trying to create my own Presto alternative that create my own Presto alternative that create my own Presto alternative that was more minimal. Ended up deciding on was more minimal. Ended up deciding on was more minimal. Ended up deciding on just using pure instead directly. I had just using pure instead directly. I had just using pure instead directly. I had it create ClaudeX which let me use my it create ClaudeX which let me use my it create ClaudeX which let me use my Codex off in Claude code so I could test Codex off in Claude code so I could test Codex off in Claude code so I could test using Claude code's workflows with using Claude code's workflows with using Claude code's workflows with Codex. And it was really cool.

  20. Codex. And it was really cool. Codex. And it was really cool. Definitely a thing I'll talk about in Definitely a thing I'll talk about in Definitely a thing I'll talk about in the near future when I start comparing the near future when I start comparing the near future when I start comparing the models more directly. Ooh, this is the models more directly. Ooh, this is the models more directly. Ooh, this is one of my favorites. I created a little one of my favorites. I created a little one of my favorites. I created a little app that is hosted on a Mac mini at my app that is hosted on a Mac mini at my app that is hosted on a Mac mini at my office that checks which Mac addresses office that checks which Mac addresses office that checks which Mac addresses are on the network to have a little are on the network to have a little are on the network to have a little tracker that shows who's currently home tracker that shows who's currently home tracker that shows who's currently home or at the office to see who's in. Makes or at the office to see who's in. Makes or at the office to see who's in. Makes it really easy to know like what to it really easy to know like what to it really easy to know like what to expect when I walk into the office. Oh, expect when I walk into the office. Oh, expect when I walk into the office. Oh, and I had it overhaul my plan service and I had it overhaul my plan service and I had it overhaul my plan service that I use for hosting all my HTML plans that I use for hosting all my HTML plans that I use for hosting all my HTML plans including the one we're looking at now. including the one we're looking at now. including the one we're looking at now. It ended up building a bunch of really It ended up building a bunch of really It ended up building a bunch of really useful stuff for it. I'll be honest useful stuff for it. I'll be honest useful stuff for it. I'll be honest though, I had Fable do the majority of though, I had Fable do the majority of though, I had Fable do the majority of the recent cleanup work cuz I only had the recent cleanup work cuz I only had the recent cleanup work cuz I only had access to Fable for a bit. But, yeah. I access to Fable for a bit. But, yeah. I access to Fable for a bit. But, yeah. I know that was a lot. I'm looking at how know that was a lot. I'm looking at how know that was a lot. I'm looking at how long I've been recording for and I'm a long I've been recording for and I'm a long I've been recording for and I'm a little scared, but I wanted to show what little scared, but I wanted to show what little scared, but I wanted to show what I've been doing with the model so that I've been doing with the model so that I've been doing with the model so that in the future when I talk about what the in the future when I talk about what the in the future when I talk about what the model is and how it works and what model is and how it works and what model is and how it works and what impressed and didn't impress me, you impressed and didn't impress me, you impressed and didn't impress me, you have the context of the things I have the context of the things I have the context of the things I actually did with it. actually did with it. actually did with it. Spoiler though, those future videos are Spoiler though, those future videos are Spoiler though, those future videos are going to be even more interesting. This going to be even more interesting. This going to be even more interesting. This model got me genuinely excited to build model got me genuinely excited to build model got me genuinely excited to build more and think bigger. It's without more and think bigger. It's without more and think bigger. It's without question one of the bigger reasons I question one of the bigger reasons I question one of the bigger reasons I gave the talk I gave recently at both gave the talk I gave recently at both gave the talk I gave recently at both Cascadia JS and AI Engineer about Cascadia JS and AI Engineer about Cascadia JS and AI Engineer about thinking bigger because this model thinking bigger because this model thinking bigger because this model pushed me to my limit. I had to push it pushed me to my limit. I had to push it pushed me to my limit. I had to push it to do bigger, harder stuff because the to do bigger, harder stuff because the to do bigger, harder stuff because the simple things were too simple for it. I simple things were too simple for it. I simple things were too simple for it. I don't have to steer it the way I did don't have to steer it the way I did don't have to steer it the way I did with 55, I just tell it go and it goes with 55, I just tell it go and it goes with 55, I just tell it go and it goes and goes and goes without needing help and goes and goes without needing help and goes and goes without needing help managing context or getting not getting managing context or getting not getting managing context or getting not getting lost. It's just kind of capable. It's a lost. It's just kind of capable. It's a lost. It's just kind of capable. It's a workhorse and I made it work hard.

  21. workhorse and I made it work hard. workhorse and I made it work hard. That said, I'm going into this without That said, I'm going into this without That said, I'm going into this without any idea how others feel and I'm curious any idea how others feel and I'm curious any idea how others feel and I'm curious how y'all do. Let me know in the how y'all do. Let me know in the how y'all do. Let me know in the comments and keep your eye out for the comments and keep your eye out for the comments and keep your eye out for the follow-up videos and again, make sure follow-up videos and again, make sure follow-up videos and again, make sure you're subbed and you hit that bell you're subbed and you hit that bell you're subbed and you hit that bell because there's going to be a lot of because there's going to be a lot of because there's going to be a lot of content about this model coming very, content about this model coming very, content about this model coming very, very soon. Thanks for sticking through very soon. Thanks for sticking through very soon. Thanks for sticking through to the end and until next time, to the end and until next time, to the end and until next time, peace, nerds.

Summary

The main theme is the extensive personal testing and usage of GPT 5.6, with a significant amount of inference performed. While no scriptures are mentioned, the topic centers on the practical capabilities and immense computational power of this new AI model. The takeaway is that GPT 5.6 can achieve a vast amount with substantial resource investment, and a detailed breakdown of its functionalities will be presented in multiple parts.

View original episode ↗