Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

opencode making a pong game in vite+react using (4bit) qwen/qwen3-235b-a22b-2507 locally, served by lmstudio. It used like 130GB of RAM, 0 issues with tool calls. This is completely usable locally now. Whether it's at claude level or not, idk yet, but I've no doubt we'll be there very soon... show more

SIGKITTEN

14,753 subscribers

30,470 views • 1 year ago •via X (Twitter)

Science & Technology Gaming Education

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

I got Llama 3 running in my browser using only my GPU with my Wi-Fi switched OFF completely client-side WebGPU is a new feature in browsers where JS can use the GPU of the device and apparently you can run LLMs on it too, and it's fast! My end goal for 🧠 is a locally run client side 100% private conversation where nothing gets sent to, processed in or stored in the cloud This will be helpful because I prefer making it web based instead of coding for 4 different native platforms (iOS, Android, MacOS, Windows) Two challenges still: 1) the user has to download the LLM model first and the smallest model is still ~3GB, but they only have to download it once 2) not all devices have fast GPUs yet for LLMs, but it's starting to become a default in most devices e.g.s smartphones

I got Llama 3 running in my browser using only my GPU with my Wi-Fi switched OFF completely client-side WebGPU is a new feature in browsers where JS can use the GPU of the device and apparently you can run LLMs on it too, and it's fast! My end goal for 🧠 is a locally run client side 100% private conversation where nothing gets sent to, processed in or stored in the cloud This will be helpful because I prefer making it web based instead of coding for 4 different native platforms (iOS, Android, MacOS, Windows) Two challenges still: 1) the user has to download the LLM model first and the smallest model is still ~3GB, but they only have to download it once 2) not all devices have fast GPUs yet for LLMs, but it's starting to become a default in most devices e.g.s smartphones

@levelsio

261,416 views • 2 years ago

One of the most exciting things about artifacts is that it's not just making existing coding jobs easier/faster, it's creating software that would never be written otherwise. Many people are now creating their first apps, with Claude's help. Niche apps that only a few people need. Internal tools for small companies. Or silly apps just for fun. Here's an example: wanted to visualize a dual monitor setup for a desk and compare how the monitors fit next to each other. Claude made a perfectly usable app for this in just a few turns! It's not mind-blowing or anything, but that's the entire point. It may not be worth coding it up yourself - but if you can have Claude make it in <5 minutes, it's quite useful.

One of the most exciting things about artifacts is that it's not just making existing coding jobs easier/faster, it's creating software that would never be written otherwise. Many people are now creating their first apps, with Claude's help. Niche apps that only a few people need. Internal tools for small companies. Or silly apps just for fun. Here's an example: wanted to visualize a dual monitor setup for a desk and compare how the monitors fit next to each other. Claude made a perfectly usable app for this in just a few turns! It's not mind-blowing or anything, but that's the entire point. It may not be worth coding it up yourself - but if you can have Claude make it in <5 minutes, it's quite useful.

Alex Albert

210,940 views • 2 years ago

Okay, I got Claude Code to actually fly my plane now It controls the keys via Chrome by running it with `claude --chrome` There is some issue with the lift vs gravity formula though so it takes a lot of speed to take off, but it's really cool to see it control and fly the plane This means it can code the game and then play it to check in a loop to theoretically improve it without my help (haven't gotten there though) I run it like this btw: `IS_SANDBOX=1 claude --dangerously-skip-permissions --chrome` (Run at your own risk)

Okay, I got Claude Code to actually fly my plane now It controls the keys via Chrome by running it with `claude --chrome` There is some issue with the lift vs gravity formula though so it takes a lot of speed to take off, but it's really cool to see it control and fly the plane This means it can code the game and then play it to check in a loop to theoretically improve it without my help (haven't gotten there though) I run it like this btw: `IS_SANDBOX=1 claude --dangerously-skip-permissions --chrome` (Run at your own risk)

@levelsio

143,316 views • 6 months ago

What Lisa Su actually held on stage: A mini PC the size of a lunchbox running Qwen3-235B locally, with no cloud and no discrete GPU Inside: the Ryzen AI Max+ 395, 128GB unified memory, 110GB usable as VRAM on Linux The first x86 chip that handles 200+ billion parameters on a single die AMD claims it beats the RTX 5080 by several times on memory-bound models — because the 5080 simply cannot fit them $1,400 to $2,500 once. cloud bills run $200 to $400 a month It pays for itself in a few months, then costs nothing per request This is not a faster GPU. it is the first real argument that your AI does not belong in someone else's data center

What Lisa Su actually held on stage: A mini PC the size of a lunchbox running Qwen3-235B locally, with no cloud and no discrete GPU Inside: the Ryzen AI Max+ 395, 128GB unified memory, 110GB usable as VRAM on Linux The first x86 chip that handles 200+ billion parameters on a single die AMD claims it beats the RTX 5080 by several times on memory-bound models — because the 5080 simply cannot fit them $1,400 to $2,500 once. cloud bills run $200 to $400 a month It pays for itself in a few months, then costs nothing per request This is not a faster GPU. it is the first real argument that your AI does not belong in someone else's data center

plutos

1,235,717 views • 1 month ago

🇺🇸🇻🇪 TRUMP: NOT AFRAID OF BOOTS ON THE GROUND IN VENEZUELA, U.S. WILL RUN COUNTRY PROPERLY "We had boots on the ground last night at a very high level, actually. We're not afraid of it. We don't mind saying it. But we're going to make sure that that country is run properly. We're not doing this in vain. We're there now. We're ready to go again if we have to. We're going to run the country right. It's going to run very judiciously, very fairly. It's going to make a lot of money. We're going to give money to the people." Source: CSPAN

🇺🇸🇻🇪 TRUMP: NOT AFRAID OF BOOTS ON THE GROUND IN VENEZUELA, U.S. WILL RUN COUNTRY PROPERLY "We had boots on the ground last night at a very high level, actually. We're not afraid of it. We don't mind saying it. But we're going to make sure that that country is run properly. We're not doing this in vain. We're there now. We're ready to go again if we have to. We're going to run the country right. It's going to run very judiciously, very fairly. It's going to make a lot of money. We're going to give money to the people." Source: CSPAN

Mario Nawfal

42,782 views • 6 months ago

ok this is wild. 10 year old gtx 1080 8gb pascal card running qwen3 8b locally at 18-20 tok/s via hermes agent and it's actually doing the thing. asked it to build a wireworld cellular automata simulator with 10 tests. autonomous run, no hand holding. expected it to fail on the tool calls. that's not what happened. write_file works. browser_navigate works. terminal commands work. file ops, package installs, version probes, environment setup. agent is firing tool calls cleanly and the model is reasoning about next steps at 18-20 tok/s. on hardware that pre dates "agentic" as a word. it even hit an npm install fail because node 12 is too old. didn't crash. didn't ask me. just started bootstrapping nvm on its own to fix the environment. 10 minutes in. 40% context used. 7.5gb of 8gb vram occupied. still going. i did not think this would work on this hardware. this is the most i've been wrong this month.

ok this is wild. 10 year old gtx 1080 8gb pascal card running qwen3 8b locally at 18-20 tok/s via hermes agent and it's actually doing the thing. asked it to build a wireworld cellular automata simulator with 10 tests. autonomous run, no hand holding. expected it to fail on the tool calls. that's not what happened. write_file works. browser_navigate works. terminal commands work. file ops, package installs, version probes, environment setup. agent is firing tool calls cleanly and the model is reasoning about next steps at 18-20 tok/s. on hardware that pre dates "agentic" as a word. it even hit an npm install fail because node 12 is too old. didn't crash. didn't ask me. just started bootstrapping nvm on its own to fix the environment. 10 minutes in. 40% context used. 7.5gb of 8gb vram occupied. still going. i did not think this would work on this hardware. this is the most i've been wrong this month.

Sudo su

92,287 views • 2 months ago

Warren Buffett: "It's better to have the talent of the whole world working on things like battery technology or whatever it may be. If we can add the resources of a whole bunch of other people to help solve our problems, I'm all for it." "If we get there first and they copy us — fine. If they get there first and we copy them — fine. It is not like they're going to be able to hoard the technology and we'll never see it." "I like to see the United States come in first on everything, but I don't think it will be terrible if somebody in China or India comes up with something that enables this country to deal with its problems." (Charlie Rose Show || 2009)

Warren Buffett: "It's better to have the talent of the whole world working on things like battery technology or whatever it may be. If we can add the resources of a whole bunch of other people to help solve our problems, I'm all for it." "If we get there first and they copy us — fine. If they get there first and we copy them — fine. It is not like they're going to be able to hoard the technology and we'll never see it." "I like to see the United States come in first on everything, but I don't think it will be terrible if somebody in China or India comes up with something that enables this country to deal with its problems." (Charlie Rose Show || 2009)

Kevin Carpenter

17,400 views • 1 month ago

🥊 Time for a new fight Qwen3.5 397B A17B vs Qwen3.6 27B 🌸 "Cherry Blossom" (↓ prompt below) Using OpenRouter for 397B Running 27B locally on a single RTX 5080 Wow! 🤯 I'm sure this one is going to divide opinions. 397B feels cinematic with the flare and shadows it added, while 27B is just beautiful; it has a completely different color palette and the leaves are falling from the branches as asked. It's really tough to decide which model is the best. I was about to call it a tie, but then I remembered that 397B needs 180GB VRAM to run locally 😅 Don't get me wrong the model is amazing, but right now the VRAM it requires doesn't feel justified. It's simply not 10× better than 27B. We'll have to dig deeper with more prompts to be sure, let's see what each one has in store for us. What do you think?

🥊 Time for a new fight Qwen3.5 397B A17B vs Qwen3.6 27B 🌸 "Cherry Blossom" (↓ prompt below) Using OpenRouter for 397B Running 27B locally on a single RTX 5080 Wow! 🤯 I'm sure this one is going to divide opinions. 397B feels cinematic with the flare and shadows it added, while 27B is just beautiful; it has a completely different color palette and the leaves are falling from the branches as asked. It's really tough to decide which model is the best. I was about to call it a tie, but then I remembered that 397B needs 180GB VRAM to run locally 😅 Don't get me wrong the model is amazing, but right now the VRAM it requires doesn't feel justified. It's simply not 10× better than 27B. We'll have to dig deeper with more prompts to be sure, let's see what each one has in store for us. What do you think?

left curve dev

21,224 views • 2 months ago

OH DEAR - Not sure I agree with doing something like this, BUT..... Is using California's rigged system against themselves acceptable.....guess we will find out. CA doesn't want anyone to even ask for any kind of ID (it's against the law) when they vote, if this is true and thousands of MAGA voters are going to show up and vote in CA whether they live there or not it should be legal, right? If illegal aliens can vote - and dead people can vote - then what's to stop MAGA voters from voting in CA whether they live there or not?

OH DEAR - Not sure I agree with doing something like this, BUT..... Is using California's rigged system against themselves acceptable.....guess we will find out. CA doesn't want anyone to even ask for any kind of ID (it's against the law) when they vote, if this is true and thousands of MAGA voters are going to show up and vote in CA whether they live there or not it should be legal, right? If illegal aliens can vote - and dead people can vote - then what's to stop MAGA voters from voting in CA whether they live there or not?

BelannF

59,950 views • 2 months ago

Orban is asking Hungarians and Europeans to renounce war with Russia. 'Go online, if necessary - look at satellite images and compare: what Ukraine looked like in 2015 and what it looks like now. You'll be shocked. And then, perhaps, we can talk about this. This is not a child's game. It's about where this is heading, to what extremes it will go, how it's approaching us and how we're getting drawn into it. If we don't respond with a refusal, we'll be tripped up, subjugated - and in the end, even with the best intentions, we'll find ourselves in a community that is striving for war, a community that calls itself the 'coalition of the willing'. - SK

Orban is asking Hungarians and Europeans to renounce war with Russia. 'Go online, if necessary - look at satellite images and compare: what Ukraine looked like in 2015 and what it looks like now. You'll be shocked. And then, perhaps, we can talk about this. This is not a child's game. It's about where this is heading, to what extremes it will go, how it's approaching us and how we're getting drawn into it. If we don't respond with a refusal, we'll be tripped up, subjugated - and in the end, even with the best intentions, we'll find ourselves in a community that is striving for war, a community that calls itself the 'coalition of the willing'. - SK

Zlatti71

103,703 views • 6 months ago

Victor Wembanyama’s French part of his post game interview: 🇫🇷: Victor, are you allowing yourself to enjoy tonight’s win or are you already thinking about the Knicks? 👽: It's impossible not to enjoy it. It's way too hard! I haven't built a large enough emotional reservoir yet—it's impossible not to enjoy a moment like this. It's overflowing. We'll have to lock back in pretty soon but for now it's an incredible feeling. #GoSpursGo #PorVida

Victor Wembanyama’s French part of his post game interview: 🇫🇷: Victor, are you allowing yourself to enjoy tonight’s win or are you already thinking about the Knicks? 👽: It's impossible not to enjoy it. It's way too hard! I haven't built a large enough emotional reservoir yet—it's impossible not to enjoy a moment like this. It's overflowing. We'll have to lock back in pretty soon but for now it's an incredible feeling. #GoSpursGo #PorVida

Wemby Alien Era

1,001,464 views • 1 month ago

Elon Musk: Education shouldn't be such a huge chore. “If you're thinking like, what is education? You're basically downloading data and algorithms into your brain. And it's actually amazingly bad in conventional education. Because it shouldn't be like this huge chore. So you're making it way, way better. But I think a lot of things that I would say you've probably heard 100 times and are fact, or if not doing, the more you can gamify the process of learning, the better. For my kids, I do not have to encourage them to play video games. I have to pry them from their hands like crack. It's like, drop that crack needle. The crack is addictive. So to the degree that you can make somehow learning like a game, but then it's better. And I think unfortunately a lot of education is very vaudevillian. You've got someone standing up there kind of lecturing at people, and they've done the same lecture 20 years in a row, and they're not very excited about it.” Interview with Sal Khan, Khan Academy, April 22, 2013.

Elon Musk: Education shouldn't be such a huge chore. “If you're thinking like, what is education? You're basically downloading data and algorithms into your brain. And it's actually amazingly bad in conventional education. Because it shouldn't be like this huge chore. So you're making it way, way better. But I think a lot of things that I would say you've probably heard 100 times and are fact, or if not doing, the more you can gamify the process of learning, the better. For my kids, I do not have to encourage them to play video games. I have to pry them from their hands like crack. It's like, drop that crack needle. The crack is addictive. So to the degree that you can make somehow learning like a game, but then it's better. And I think unfortunately a lot of education is very vaudevillian. You've got someone standing up there kind of lecturing at people, and they've done the same lecture 20 years in a row, and they're not very excited about it.” Interview with Sal Khan, Khan Academy, April 22, 2013.

ELON CLIPS

34,228 views • 1 month ago

Boris Cherny, Head of Claude Code at Anthropic: "I do not prompt Claude anymore. I have loops running that prompt Claude. My job is to write loops." The person who built the most used AI coding tool stopped prompting entirely. Most people using it every day are still typing one request at a time. This Github repo changes that. ↳ Seven production loop patterns with real token cost estimates ↳ CLI that scaffolds your entire loop in five minutes ↳ Starters for Grok, Claude Code, Codex, and Opencode The repo gives you the framework. My article covers the mental shift that makes it actually work.

Boris Cherny, Head of Claude Code at Anthropic: "I do not prompt Claude anymore. I have loops running that prompt Claude. My job is to write loops." The person who built the most used AI coding tool stopped prompting entirely. Most people using it every day are still typing one request at a time. This Github repo changes that. ↳ Seven production loop patterns with real token cost estimates ↳ CLI that scaffolds your entire loop in five minutes ↳ Starters for Grok, Claude Code, Codex, and Opencode The repo gives you the framework. My article covers the mental shift that makes it actually work.

Kanika

20,148 views • 2 days ago

I'm thrilled to announce the definitive course on Claude Code, created with Anthropic and taught by Elie Schoppik Elie Schoppik. If you want to use highly agentic coding - where AI works autonomously for many minutes or longer, not just completing code snippets - this is it. Claude Code has been a game-changer for many developers (including me!), but there's real depth to using it well. This comprehensive course covers everything from fundamentals to advanced patterns. After this short course, you'll be able to: - Orchestrate multiple Claude subagents to work on different parts of your codebase simultaneously - Tag Claude in GitHub issues and have it autonomously create, review, and merge pull requests - Transform messy Jupyter notebooks into clean, production-ready dashboards - Use MCP tools like Playwright so Claude can see what's wrong with your UI and fix it autonomously Whether you're new to Claude Code or already using it, you'll discover powerful capabilities that can fundamentally change how you build software. I'm very excited about what agentic coding lets everyone now do. Please take this course!

I'm thrilled to announce the definitive course on Claude Code, created with Anthropic and taught by Elie Schoppik Elie Schoppik. If you want to use highly agentic coding - where AI works autonomously for many minutes or longer, not just completing code snippets - this is it. Claude Code has been a game-changer for many developers (including me!), but there's real depth to using it well. This comprehensive course covers everything from fundamentals to advanced patterns. After this short course, you'll be able to: - Orchestrate multiple Claude subagents to work on different parts of your codebase simultaneously - Tag Claude in GitHub issues and have it autonomously create, review, and merge pull requests - Transform messy Jupyter notebooks into clean, production-ready dashboards - Use MCP tools like Playwright so Claude can see what's wrong with your UI and fix it autonomously Whether you're new to Claude Code or already using it, you'll discover powerful capabilities that can fundamentally change how you build software. I'm very excited about what agentic coding lets everyone now do. Please take this course!

Andrew Ng

1,648,421 views • 11 months ago

I've seen those WhatsApp group msgs on banning us and honestly we got 99 problems and they aren't one. You see , The main purpose of a pre-purchase inspection is not to tell a client whether to buy a car or not, but to inform them of the exact condition of the unit they intend to purchase. In fact, we make it a point to help clients especially first-time buyers set realistic expectations. You can’t expect a 2010 locally used car to be in the same condition as a fresh import. However, one thing we absolutely refuse to do is lie to clients. If a car is mixing oil with coolant, we say it as it is. If the mileage has been tampered with or the car is full of rust, we call it out. And if any dealer has a problem with that, we honestly couldn’t care less. SORRY NOT SORRY. Anyways, for pre purchase inspections, WhatsApp us on 0729686646.

I've seen those WhatsApp group msgs on banning us and honestly we got 99 problems and they aren't one. You see , The main purpose of a pre-purchase inspection is not to tell a client whether to buy a car or not, but to inform them of the exact condition of the unit they intend to purchase. In fact, we make it a point to help clients especially first-time buyers set realistic expectations. You can’t expect a 2010 locally used car to be in the same condition as a fresh import. However, one thing we absolutely refuse to do is lie to clients. If a car is mixing oil with coolant, we say it as it is. If the mileage has been tampered with or the car is full of rust, we call it out. And if any dealer has a problem with that, we honestly couldn’t care less. SORRY NOT SORRY. Anyways, for pre purchase inspections, WhatsApp us on 0729686646.

4Real

67,337 views • 9 months ago

This video is from a cemetery in Mariupol. It's a long video. It's a video that's very difficult to watch but it must be seen by everyone. Especially by politicians all over the world who are deciding whether or not to continue supporting Ukraine’s fight for its right to exist. What you see in this video is the future fate of all Ukrainian cities if Russia is not stopped. #Ukraine #Mariupol #StandWithUkraine #Russia #RussiaUkraineWar #RussiaIsATerroristState

This video is from a cemetery in Mariupol. It's a long video. It's a video that's very difficult to watch but it must be seen by everyone. Especially by politicians all over the world who are deciding whether or not to continue supporting Ukraine’s fight for its right to exist. What you see in this video is the future fate of all Ukrainian cities if Russia is not stopped. #Ukraine #Mariupol #StandWithUkraine #Russia #RussiaUkraineWar #RussiaIsATerroristState

Natalka

127,447 views • 2 years ago

🚨 JUST IN: Sen. John Kennedy just said it PERFECTLY "NOTHING that the Trump administration does, no matter what it is, even if it's something very, very, very positive, he's not given CREDIT for it by many members of the media. That's AGGRAVATING." 💯 "I've said this before, but it's frustrating sometimes that Washington, D.C., you know it, it's a Democratic town, not just the Democrats in Congress, but the think tanks, the academics, the bureaucrats, and that can be frustrating." H/t TV News Now

🚨 JUST IN: Sen. John Kennedy just said it PERFECTLY "NOTHING that the Trump administration does, no matter what it is, even if it's something very, very, very positive, he's not given CREDIT for it by many members of the media. That's AGGRAVATING." 💯 "I've said this before, but it's frustrating sometimes that Washington, D.C., you know it, it's a Democratic town, not just the Democrats in Congress, but the think tanks, the academics, the bureaucrats, and that can be frustrating." H/t TV News Now

Eric Daugherty

111,696 views • 7 months ago

Muslim TRIES TO CORNER Christian... But IT COMPLETELY BACKFIRES Right now, somewhere in the world, someone is hearing about Jesus for the very first time. Not millions. One person. With a name. With a story. With a life that will never be the same. You're watching this for a reason. ───────────────────────────────────── Help Support Dr. Raf Show: Relevant Articles / Sources: Full Clip: (This channel is not run by those associated with the clips). ───────────────────────────────────── HELP REACH THE UNREACHED Partner Monthly: Join the Channel: God Bless you for standing with this mission 🙏 #Islam #Quran #Muslim

Muslim TRIES TO CORNER Christian... But IT COMPLETELY BACKFIRES Right now, somewhere in the world, someone is hearing about Jesus for the very first time. Not millions. One person. With a name. With a story. With a life that will never be the same. You're watching this for a reason. ───────────────────────────────────── Help Support Dr. Raf Show: Relevant Articles / Sources: Full Clip: (This channel is not run by those associated with the clips). ───────────────────────────────────── HELP REACH THE UNREACHED Partner Monthly: Join the Channel: God Bless you for standing with this mission 🙏 #Islam #Quran #Muslim

Towards Jesus

14,167 views • 1 month ago

𝗢𝗹𝗹𝗮𝗺𝗮 + 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗷𝘂𝘀𝘁 𝗺𝗮𝗱𝗲 𝗔𝗜 𝗰𝗼𝗱𝗶𝗻𝗴 𝗳𝗿𝗲𝗲. 𝗔𝗻𝗱 𝗶𝘁 𝗿𝘂𝗻𝘀 𝗰𝗼𝗺𝗽𝗹𝗲𝘁𝗲𝗹𝘆 𝗹𝗼𝗰𝗮𝗹. No subscriptions. No limits. No data leaving your machine. Claude Code used to cost $3-$15 per million tokens. Now you can run it with free open-source models on your computer. January 16th, 2026: Ollama version 0.14.0 became compatible with Anthropic's messages API. That means Claude Code now works with any Ollama model. Zero ongoing costs. Your code never leaves your computer. Here's the setup: Download Ollama. Run `ollama pull qwen3-coder`. Install Claude Code. Set two environment variables to connect them. Then run: `claude --model qwen3-coder` Give it tasks in plain English: "Write a function to process images." "Debug why my API calls are failing." It reads your files. Makes changes. Tests the code. Shows you what it did. Top models: Qwen 3-Coder for everyday tasks. GPT-OSS 20B for complex projects. This is completely legal. Explicitly supported by both tools.

𝗢𝗹𝗹𝗮𝗺𝗮 + 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗷𝘂𝘀𝘁 𝗺𝗮𝗱𝗲 𝗔𝗜 𝗰𝗼𝗱𝗶𝗻𝗴 𝗳𝗿𝗲𝗲. 𝗔𝗻𝗱 𝗶𝘁 𝗿𝘂𝗻𝘀 𝗰𝗼𝗺𝗽𝗹𝗲𝘁𝗲𝗹𝘆 𝗹𝗼𝗰𝗮𝗹. No subscriptions. No limits. No data leaving your machine. Claude Code used to cost $3-$15 per million tokens. Now you can run it with free open-source models on your computer. January 16th, 2026: Ollama version 0.14.0 became compatible with Anthropic's messages API. That means Claude Code now works with any Ollama model. Zero ongoing costs. Your code never leaves your computer. Here's the setup: Download Ollama. Run `ollama pull qwen3-coder`. Install Claude Code. Set two environment variables to connect them. Then run: `claude --model qwen3-coder` Give it tasks in plain English: "Write a function to process images." "Debug why my API calls are failing." It reads your files. Makes changes. Tests the code. Shows you what it did. Top models: Qwen 3-Coder for everyday tasks. GPT-OSS 20B for complex projects. This is completely legal. Explicitly supported by both tools.

Julian Goldie SEO

168,609 views • 6 months ago