Ziwen

@ziwenxu_ • 19,005 subscribers

26 | Trying new AI tools, building things, and sharing what’s worth your time. https://t.co/uuEn9gTnE0 https://t.co/X7SBufnpNs

Shorts

Day 1 of building GTA 6. Still feels fake typing that out. Upgraded to Claude Max 20x just for this. Spent a couple hours getting the whole project structured and pushed to the repo. Sandbox is up and running. No studio, no publisher. Just whoever shows up. We picked Godot on purpose: it's community-owned, so nobody can pull an EPIC on us later and rewrite the deal once we're invested. The goal: beat the real GTA 6 to launch. Ambitious, probably stupid, doing it anyway. If you can model, code, build levels, or write music and lore, come join. Looking for a couple contributors to cook this. Time to cook. Reply if you're in.

Day 1 of building GTA 6. Still feels fake typing that out. Upgraded to Claude Max 20x just for this. Spent a couple hours getting the whole project structured and pushed to the repo. Sandbox is up and running. No studio, no publisher. Just whoever shows up. We picked Godot on purpose: it's community-owned, so nobody can pull an EPIC on us later and rewrite the deal once we're invested. The goal: beat the real GTA 6 to launch. Ambitious, probably stupid, doing it anyway. If you can model, code, build levels, or write music and lore, come join. Looking for a couple contributors to cook this. Time to cook. Reply if you're in.

6,088,886 views

OpenCode Go is now wired into Codex!! The pricing is insane. $10 gets you 10,000 DeepSeek requests every 5 hours (no weekly limits). I converted that into DeepSeek API dollars because I thought I was reading it wrong, and the same 5 hours of usage would run somewhere between $10 and $30 depending on how big your context gets. So one afternoon of use already covers the whole sub. It comes with Kimi K3 as well. Both sit in the picker next to my other models now. Next time we hit a limit in the middle of a loop, we can grab it and keep going.

OpenCode Go is now wired into Codex!! The pricing is insane. $10 gets you 10,000 DeepSeek requests every 5 hours (no weekly limits). I converted that into DeepSeek API dollars because I thought I was reading it wrong, and the same 5 hours of usage would run somewhere between $10 and $30 depending on how big your context gets. So one afternoon of use already covers the whole sub. It comes with Kimi K3 as well. Both sit in the picker next to my other models now. Next time we hit a limit in the middle of a loop, we can grab it and keep going.

435,937 views

Codex can run Qwen-3.8-max now as well!! Alibaba most capable model, dropped today and it's already in my codex picker. It's a token plan subscription, not metered api billing. You take the key from your Qwen plan, drop it into Codex Router, and it spends down the plan instead of your card. There's a catch though. Qwen's official setup switches your whole codex over to them, so your ChatGPT models stop showing up at all. That's exactly what Codex Router is for. It adds models to the list instead of replacing them, so sol, Grok, kimi, Deepseek and now Qwen 3.8 max all sit in the same picker and it can grab whichever one suits the job. Router's open source, setup's in the video 👇

Codex can run Qwen-3.8-max now as well!! Alibaba most capable model, dropped today and it's already in my codex picker. It's a token plan subscription, not metered api billing. You take the key from your Qwen plan, drop it into Codex Router, and it spends down the plan instead of your card. There's a catch though. Qwen's official setup switches your whole codex over to them, so your ChatGPT models stop showing up at all. That's exactly what Codex Router is for. It adds models to the list instead of replacing them, so sol, Grok, kimi, Deepseek and now Qwen 3.8 max all sit in the same picker and it can grab whichever one suits the job. Router's open source, setup's in the video 👇

419,031 views

Union Alpha is now in Codex!! The speed is insane over 300 to 400 tokens a second, 262k context, images in, and it's free for a week. It's a good timing too, most of the Codex usage is basically gone right now. Zai did this exact thing in August: Ox Alpha showed up unnamed, free for a week, built for agentic coding, and a week later it was GLM-5.3-Flash. So what model do you guys think it is?

Union Alpha is now in Codex!! The speed is insane over 300 to 400 tokens a second, 262k context, images in, and it's free for a week. It's a good timing too, most of the Codex usage is basically gone right now. Zai did this exact thing in August: Ox Alpha showed up unnamed, free for a week, built for agentic coding, and a week later it was GLM-5.3-Flash. So what model do you guys think it is?

99,191 views

We can now run GPT-5.6 Luna Max over 5,000 extra times a week for only $10. A lot of us are out of Codex usage right now. So we wired Luna Max from OpenCode Go directly into our Codex Router!! It's the same Luna we already run in Codex, it just spends the $10 sub instead of our ChatGPT plan, and the reasoning level still works so we keep it on max. The caps are 5,100 Luna requests per week and 10,250 per month. Codex cuts us off, we've still got a full week of Luna sitting there. And we never leave Codex. just one more model in the picker, grab it when the wall hits and keep going.

We can now run GPT-5.6 Luna Max over 5,000 extra times a week for only $10. A lot of us are out of Codex usage right now. So we wired Luna Max from OpenCode Go directly into our Codex Router!! It's the same Luna we already run in Codex, it just spends the $10 sub instead of our ChatGPT plan, and the reasoning level still works so we keep it on max. The caps are 5,100 Luna requests per week and 10,250 per month. Codex cuts us off, we've still got a full week of Luna sitting there. And we never leave Codex. just one more model in the picker, grab it when the wall hits and keep going.

153,122 views

Codex can now run Deepseek-v4- flash! There's a catch though. Deepseek's official setup switches your entire codex over to them, so your GPT models stop showing up at all. This is exactly what Codex Router is for. It adds models to the list instead of replacing them, so sol, grok, kimi and deepseek all sit in the same picker and i just grab whichever one suits the job. Deepseek v4-flash is $0.28 per million output tokens. opus 4.8 is $25. same picker, 89x apart. Links in the comment. setup's in the video 👇

Codex can now run Deepseek-v4- flash! There's a catch though. Deepseek's official setup switches your entire codex over to them, so your GPT models stop showing up at all. This is exactly what Codex Router is for. It adds models to the list instead of replacing them, so sol, grok, kimi and deepseek all sit in the same picker and i just grab whichever one suits the job. Deepseek v4-flash is $0.28 per million output tokens. opus 4.8 is $25. same picker, 89x apart. Links in the comment. setup's in the video 👇

145,018 views

Tested Fugu-ultra on the questions that break most of the AI: - how many r's in strawberry - is 5.11 bigger than 5.1 - the famous car/wash question (Twice) I think we just got Fable Back

Tested Fugu-ultra on the questions that break most of the AI: - how many r's in strawberry - is 5.11 bigger than 5.1 - the famous car/wash question (Twice) I think we just got Fable Back

188,692 views

AAA studios spend $200M on a city this good. Ours cost $1,500 and swapped Miami for SF within a week. An agent did it in a loop while we slept, now he's working on the traffic and pedestrian.

AAA studios spend $200M on a city this good. Ours cost $1,500 and swapped Miami for SF within a week. An agent did it in a loop while we slept, now he's working on the traffic and pedestrian.

171,663 views

Muse Spark 1.3 is now in Codex!! OpenCode's free tier runs it for nothing, and $10 on OpenCode Go gets you over 45,300 requests. It's basically Opus 5 tier too. There's a catch though. Meta may use your inputs and outputs for training, which is the whole reason the tier is called contributor and the whole reason it's free. Muse 1.3 just sits in the picker next to my paid models now. I'll just use Astra + Muse Spark for subagents. Gonna keep everything in stock in case we need it.

Muse Spark 1.3 is now in Codex!! OpenCode's free tier runs it for nothing, and $10 on OpenCode Go gets you over 45,300 requests. It's basically Opus 5 tier too. There's a catch though. Meta may use your inputs and outputs for training, which is the whole reason the tier is called contributor and the whole reason it's free. Muse 1.3 just sits in the picker next to my paid models now. I'll just use Astra + Muse Spark for subagents. Gonna keep everything in stock in case we need it.

92,093 views

Grok 4.5 is now in Codex now too. No API key, no billing setup, no config file. If you have X Premium, you sign in with your X account and Grok 4.5 runs inside your Codex. OAuth handles it. Your subscription is your access. That's it. And Grok 4.5 just got a lot better, so if you already pay for X Premium, you can now run it in Codex for free.

Grok 4.5 is now in Codex now too. No API key, no billing setup, no config file. If you have X Premium, you sign in with your X account and Grok 4.5 runs inside your Codex. OAuth handles it. Your subscription is your access. That's it. And Grok 4.5 just got a lot better, so if you already pay for X Premium, you can now run it in Codex for free.

111,591 views

A loop of AI agents built me a gun. It just can't make it shoot. Day 11 of building GTA 6 with a loop of agents. Yesterday I said today they'd learn to steal cars and shoot. Today's progress: - 2 weapons in - Pickup system works (we can grab weapons now) The downside: you pull the trigger, nothing happens. The loop is incredible at the overall, but shooting is the main part of the game, so I think we wire this one up ourselves. So here is the progress I'm taking shooting and sitting down to learn the physics myself with agent as support. At the same time J A Z I I takes driving by hand. The agents keep doing what they're good at. They're building out the police system and the economy right now, while I type this. Also spent the weekend in SF at a Tripo3D Donut event learning 3D art.

A loop of AI agents built me a gun. It just can't make it shoot. Day 11 of building GTA 6 with a loop of agents. Yesterday I said today they'd learn to steal cars and shoot. Today's progress: - 2 weapons in - Pickup system works (we can grab weapons now) The downside: you pull the trigger, nothing happens. The loop is incredible at the overall, but shooting is the main part of the game, so I think we wire this one up ourselves. So here is the progress I'm taking shooting and sitting down to learn the physics myself with agent as support. At the same time J A Z I I takes driving by hand. The agents keep doing what they're good at. They're building out the police system and the economy right now, while I type this. Also spent the weekend in SF at a Tripo3D Donut event learning 3D art.

67,385 views

My whole city is walking around with its arms stretched wide open and I can't get them to put them down. day 12 of building GTA 6 with a loop of AI agents. one of them spawned a crowd overnight and forgot to attach the animation. so now everyone walks around arms-out like a town full of scarecrows. Good news, weapon holding animations finally work. pick a gun up off the ground and it actually sits in your hands now. Today I'm going to work on the shooting, by hand, like I said I would yesterday. recoil, the round leaving the barrel, the hit actually landing. Can't wait to start shooting!!

My whole city is walking around with its arms stretched wide open and I can't get them to put them down. day 12 of building GTA 6 with a loop of AI agents. one of them spawned a crowd overnight and forgot to attach the animation. so now everyone walks around arms-out like a town full of scarecrows. Good news, weapon holding animations finally work. pick a gun up off the ground and it actually sits in your hands now. Today I'm going to work on the shooting, by hand, like I said I would yesterday. recoil, the round leaving the barrel, the hit actually landing. Can't wait to start shooting!!

59,726 views

Grok 4.7 is lazy. I ran it in Cursor for 10 hours, and I gave it the image from every angle. Sad to say the hype is not worth it. Not good enough just yet. The launch benchmarks say the same thing, 26% on Terminal-Bench 4.0. Have to see what Grok 5 looks like.

Grok 4.7 is lazy. I ran it in Cursor for 10 hours, and I gave it the image from every angle. Sad to say the hype is not worth it. Not good enough just yet. The launch benchmarks say the same thing, 26% on Terminal-Bench 4.0. Have to see what Grok 5 looks like.

10,420 views

I couldn't stop checking if Fable was back. So I made it check itself.

I couldn't stop checking if Fable was back. So I made it check itself.

19,642 views

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

The first ever CLAUDE MADE ANIME is out!! It's called Gus & Spark and 25 minute episode. Opus 5.5 MAX worked on it for over 15 hours, and wrote it code by code with 0 AI generated video. - 34K lines of code - 986M tokens, about $370 at API prices I believe Fable 5.5 should be able to produce even higher quality anime! In the future all 3D animated videos can be created by Claude without any AI generated video.

The first ever CLAUDE MADE ANIME is out!! It's called Gus & Spark and 25 minute episode. Opus 5.5 MAX worked on it for over 15 hours, and wrote it code by code with 0 AI generated video. - 34K lines of code - 986M tokens, about $370 at API prices I believe Fable 5.5 should be able to produce even higher quality anime! In the future all 3D animated videos can be created by Claude without any AI generated video.

365,991 views • 23 days ago

Claude has a big privacy problem. Shared conversations are showing up in search results. I can search Brave right now and read strangers’ chat histories, names, work, whatever they pasted in. Anthropic this needs fixing today.

Claude has a big privacy problem. Shared conversations are showing up in search results. I can search Brave right now and read strangers’ chat histories, names, work, whatever they pasted in. Anthropic this needs fixing today.

737,508 views • 3 months ago

I just one-shotted Minecraft with Claude Fable. One prompt. No edits. No follow-ups. Blocks, terrain, building, breaking. All working. Cost me 10% of my 5-hour usage window. We've officially run out of excuses to not build things.

I just one-shotted Minecraft with Claude Fable. One prompt. No edits. No follow-ups. Blocks, terrain, building, breaking. All working. Cost me 10% of my 5-hour usage window. We've officially run out of excuses to not build things.

737,913 views • 5 months ago

Introducing OpenAI "o", your personal assistant!! It showed up on the ChatGPT Pro upgrade screen as "o, your always-on assistant". Always-on means it keeps working after you close the chat. ppl say it runs on a new model called aeon, built for long tasks. Same lane as Grok Bot. I'm already on Pro, so if the leak is right it lands straight in my sub. OpenAI's DevDay is Tuesday. Can't wait to test it out!!

Introducing OpenAI "o", your personal assistant!! It showed up on the ChatGPT Pro upgrade screen as "o, your always-on assistant". Always-on means it keeps working after you close the chat. ppl say it runs on a new model called aeon, built for long tasks. Same lane as Grok Bot. I'm already on Pro, so if the leak is right it lands straight in my sub. OpenAI's DevDay is Tuesday. Can't wait to test it out!!

166,211 views • 1 month ago

Ox Alpha is wired into codex now, and it's free for the whole week!! no api key, no billing. It sits in the picker next to the models we already pay for, and our native models never moved. Vision and a million tokens of context, costing nothing until the window shuts. so we stopped picking the files we think matter and started handing it the whole repo. Seven days is the whole deal. after that it goes back to being something we'd have to pay for, so this week is for the jobs that are usually too big. Here's the whole setup👇

Ox Alpha is wired into codex now, and it's free for the whole week!! no api key, no billing. It sits in the picker next to the models we already pay for, and our native models never moved. Vision and a million tokens of context, costing nothing until the window shuts. so we stopped picking the files we think matter and started handing it the whole repo. Seven days is the whole deal. after that it goes back to being something we'd have to pay for, so this week is for the jobs that are usually too big. Here's the whole setup👇

271,047 views • 2 months ago

Rockstar just announced GTA 6 preorders open June 25. We are at day 9 of building my own with a loop of AI agents, and the streets finally have people in them. The map's set, so I'm leaving it as it is and slowly adding more stuff into it. the honest part: everyone's hyping Unreal's new 5.8 update and its official MCP. I tried it for whole night. My feedback is not to update it yet. The 5.8 MCP isn't like crazy good, and it actually slows the build down. Chasing the newest version felt like the obvious move. it wasn't. the loop runs better without it. Repo's still open, the project just lives in Unreal now. Theirs is a preorder. mine you can pull and run yourself.

Rockstar just announced GTA 6 preorders open June 25. We are at day 9 of building my own with a loop of AI agents, and the streets finally have people in them. The map's set, so I'm leaving it as it is and slowly adding more stuff into it. the honest part: everyone's hyping Unreal's new 5.8 update and its official MCP. I tried it for whole night. My feedback is not to update it yet. The 5.8 MCP isn't like crazy good, and it actually slows the build down. Chasing the newest version felt like the obvious move. it wasn't. the loop runs better without it. Repo's still open, the project just lives in Unreal now. Theirs is a preorder. mine you can pull and run yourself.

384,200 views • 4 months ago

Shooting, aiming, and the animations are all done. Day 17 of speed running to build the GTA 6 with a loop of AI agents. Today's progress: - Shooting feels real now, the hits actually land. - Animations are actually connected, the character moves like a real game instead of a cheap asset. Remember the animation wall from the last two days? We bought a premium pack off Epic, handed it to Claude, and the character finally moves right. The whole on foot half is basically there. What's left for the demo is driving and a multiplayer mode. Aiming to finish the whole thing this week so everyone can actually play it. J A Z I I can't wait to implement driving and see the whole game.

Shooting, aiming, and the animations are all done. Day 17 of speed running to build the GTA 6 with a loop of AI agents. Today's progress: - Shooting feels real now, the hits actually land. - Animations are actually connected, the character moves like a real game instead of a cheap asset. Remember the animation wall from the last two days? We bought a premium pack off Epic, handed it to Claude, and the character finally moves right. The whole on foot half is basically there. What's left for the demo is driving and a multiplayer mode. Aiming to finish the whole thing this week so everyone can actually play it. J A Z I I can't wait to implement driving and see the whole game.

357,465 views • 4 months ago

Watch how I got the Poke Pro plan for $0.

Watch how I got the Poke Pro plan for $0.

392,004 views • 5 months ago

It worked!! Kimi k3 Oauth is now wired directly into codex. The Kimi k3 model is insane but their app is not. So I figure a better way to use k3 by hooking my sub straight into codex instead. K3 now sits in the model picker next to Sol 5.6. Same subscription i was already paying for, no api key, just have to do OAuth in Kimi Code. Links down below 👇

It worked!! Kimi k3 Oauth is now wired directly into codex. The Kimi k3 model is insane but their app is not. So I figure a better way to use k3 by hooking my sub straight into codex instead. K3 now sits in the model picker next to Sol 5.6. Same subscription i was already paying for, no api key, just have to do OAuth in Kimi Code. Links down below 👇

277,302 views • 3 months ago

AI can't replace a 3D artist yet. I gave a loop of AI agents two days to fix my animations. This is what I got back. Day 15 + 16 of building my own GTA 6 with a loop of AI agents. Today's progress: - Aiming works now, you can line up a shot properly. - Camera movement is in, it follows you the way it should. - Walking turns look right, no more sliding around corners. We have spent two whole days letting the loop work on the animations and got basically no improvement. The ones we're ending up with and the free ones off Mixamo both look cheap. So we're probably just going to buy a premium animation pack off Epic and hand it to Claude. Good news is aiming and the camera feel real now. Making the character move like it isn't a cheap asset is the next wall.

AI can't replace a 3D artist yet. I gave a loop of AI agents two days to fix my animations. This is what I got back. Day 15 + 16 of building my own GTA 6 with a loop of AI agents. Today's progress: - Aiming works now, you can line up a shot properly. - Camera movement is in, it follows you the way it should. - Walking turns look right, no more sliding around corners. We have spent two whole days letting the loop work on the animations and got basically no improvement. The ones we're ending up with and the free ones off Mixamo both look cheap. So we're probably just going to buy a premium animation pack off Epic and hand it to Claude. Good news is aiming and the camera feel real now. Making the character move like it isn't a cheap asset is the next wall.

320,498 views • 4 months ago

Day 2 of building my GTA 6 agent in the loop. It's working better than yesterday, genuinely. But two things are bugging me. The Godot version still looks cheap, or might be because half of it isn't finished. The agent built downtown LA skyscrapers, which is a problem, because this is supposed to be Florida. Also I've burned 33% of my 20x weekly usage in one day. So that clock is ticking. So here's where I'm stuck: switch to Unreal, or keep grinding on Godot?

Day 2 of building my GTA 6 agent in the loop. It's working better than yesterday, genuinely. But two things are bugging me. The Godot version still looks cheap, or might be because half of it isn't finished. The agent built downtown LA skyscrapers, which is a problem, because this is supposed to be Florida. Also I've burned 33% of my 20x weekly usage in one day. So that clock is ticking. So here's where I'm stuck: switch to Unreal, or keep grinding on Godot?

345,035 views • 5 months ago

IT WORKED!! Here's how you run Claude Code + Local LLM on your own machine in under 90 seconds

IT WORKED!! Here's how you run Claude Code + Local LLM on your own machine in under 90 seconds

467,415 views • 7 months ago

I let a loop of AI agents build GTA 6 while I slept. Woke up to a dead Mac and a half walking NPC. Day 8. We're moving the whole game from Godot to Unreal, and the agents kept shipping right through it: - the first NPC walks but no animations - in-game phone now has maps, contacts, and a wallet. Bad news: the Mac died overnight because every agent opened the game at once to screenshot it. Fix was one line in claude.md, only one game or editor open at a time. Agents are back to looping, one at a time. Tomorrow: finish converting the entire repo to Unreal, and work the walk animation so the NPC actually looks real. The loop keeps building the phone apps, and I might pull the map onto the phone so the whole thing feels seamless. One catch: the licensed Unreal assets can't be shared publicly, so that part stays in a private build.

I let a loop of AI agents build GTA 6 while I slept. Woke up to a dead Mac and a half walking NPC. Day 8. We're moving the whole game from Godot to Unreal, and the agents kept shipping right through it: - the first NPC walks but no animations - in-game phone now has maps, contacts, and a wallet. Bad news: the Mac died overnight because every agent opened the game at once to screenshot it. Fix was one line in claude.md, only one game or editor open at a time. Agents are back to looping, one at a time. Tomorrow: finish converting the entire repo to Unreal, and work the walk animation so the NPC actually looks real. The loop keeps building the phone apps, and I might pull the map onto the phone so the whole thing feels seamless. One catch: the licensed Unreal assets can't be shared publicly, so that part stays in a private build.

290,991 views • 4 months ago

10 days ago this was an empty map. A loop of AI agents has been building GTA 6 on it and the community building it together while I sleep. day 10. look at it now. People are walking around. a skyline that actually holds up from the air. from a distance you'd swear it was the real one. Unreal is so goated, this would've taken months on Godot. Up close it's still rough, not every building can enter. on night time the far side of the ocean is empty, so the water goes black. might build a second city over there to fix it. still deciding. Going to implement steal cars and shoot tomorrow. Any suggestions?

10 days ago this was an empty map. A loop of AI agents has been building GTA 6 on it and the community building it together while I sleep. day 10. look at it now. People are walking around. a skyline that actually holds up from the air. from a distance you'd swear it was the real one. Unreal is so goated, this would've taken months on Godot. Up close it's still rough, not every building can enter. on night time the far side of the ocean is empty, so the water goes black. might build a second city over there to fix it. still deciding. Going to implement steal cars and shoot tomorrow. Any suggestions?

210,499 views • 4 months ago

Command Code is now in Codex!! The plan we're on is $1 a month. It comes with $10 of credits and around 15K requests. GPT-5.6 Luna is in it, the same model we just moved to when the OpenCode sub got cut, and Grok 4.5 sits in there too next to DeepSeek, Kimi, GLM and MiniMax. Their own docs say the $1 tier has no API access. It's in our picker anyway. So there's a dollar sub in there next to everything else now. Next time a limit hits mid loop we grab it and keep going.

Command Code is now in Codex!! The plan we're on is $1 a month. It comes with $10 of credits and around 15K requests. GPT-5.6 Luna is in it, the same model we just moved to when the OpenCode sub got cut, and Grok 4.5 sits in there too next to DeepSeek, Kimi, GLM and MiniMax. Their own docs say the $1 tier has no API access. It's in our picker anyway. So there's a dollar sub in there next to everything else now. Next time a limit hits mid loop we grab it and keep going.

122,460 views • 2 months ago

We just one shotted Stardew Valley in 54 hours with Opus 5.5 It's crazy to even say it but it has its own storyline, every single feature we could imagine, including, crafting, planting, fishing, even the vibe itself. 12 agents ran in parallel the whole time, one per area of the game. each one builds, gets torn apart by a critic. I can say Opus 5.5 is the true goat. IMO It can truly surpass Stardew Valley if the loop finishes and even publish into steam right the way as demo. By the way we only spent $200 on this. The original creator spent years to get to this point.

We just one shotted Stardew Valley in 54 hours with Opus 5.5 It's crazy to even say it but it has its own storyline, every single feature we could imagine, including, crafting, planting, fishing, even the vibe itself. 12 agents ran in parallel the whole time, one per area of the game. each one builds, gets torn apart by a critic. I can say Opus 5.5 is the true goat. IMO It can truly surpass Stardew Valley if the loop finishes and even publish into steam right the way as demo. By the way we only spent $200 on this. The original creator spent years to get to this point.

69,355 views • 1 month ago

Day 3 of building GTA 6 with a loop of AI agents. We've got: - NPCs walking around - cars on the road - shadows and reflections that finally look right - weapons - phones with Instagram in it wait. imagine we wire those in-game phones to real X accounts. You drive past someone and their actual timeline is sitting on their phone. The agents are still looping. Tell me what features you guys want I will add it in there

Day 3 of building GTA 6 with a loop of AI agents. We've got: - NPCs walking around - cars on the road - shadows and reflections that finally look right - weapons - phones with Instagram in it wait. imagine we wire those in-game phones to real X accounts. You drive past someone and their actual timeline is sitting on their phone. The agents are still looping. Tell me what features you guys want I will add it in there

207,491 views • 5 months ago

Many of us burned over 80% of our Codex usage in a single day. Stop whatever you're running right now. Astra at the top is what's eating it. Every plan, every review, all day, on the most expensive model you have. So let's flip it. Put DeepSeek V4.1 Flash on top as the orchestrator by running off OpenCode Go or Command Code connect it into codex then spawn Astra as a subagent only for the one task that actually needs it. Let's max out of what's left.

Many of us burned over 80% of our Codex usage in a single day. Stop whatever you're running right now. Astra at the top is what's eating it. Every plan, every review, all day, on the most expensive model you have. So let's flip it. Put DeepSeek V4.1 Flash on top as the orchestrator by running off OpenCode Go or Command Code connect it into codex then spawn Astra as a subagent only for the one task that actually needs it. Let's max out of what's left.

69,260 views • 1 month ago

We tripled the fps in one day everyone!! 15 yesterday, 40 to 50 today. Day 32 of building our own community open source, open world game with a loop of AI agents. the whole day went into performance and it paid off next stop is locking 60

We tripled the fps in one day everyone!! 15 yesterday, 40 to 50 today. Day 32 of building our own community open source, open world game with a loop of AI agents. the whole day went into performance and it paid off next stop is locking 60

134,808 views • 3 months ago

Someone put the before and after Tibo in the same room. This is better than my reply

Someone put the before and after Tibo in the same room. This is better than my reply

84,877 views • 2 months ago