Loading video...

Video Failed to Load

Go Home

Big win for open-source LLMs! DeepSeek V4 Pro holds the top open-weights score on SWE-bench Verified, in the GPT-5.5 range. GLM 5.2 leads the open-weight intelligence index and sits near the closed frontier on long-horizon coding. But this leaderboard number is a weak proxy for real performance. It comes...

_avichawla's profile picture

Avi Chawla

72,646 subscribers

44,124 views • 1 month ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Don't train the model, evolve the harness. I read a brilliant blog post from Hugging Face where they took a frozen open model scoring 0% on a hard legal agent benchmark, left its weights alone, and let an automated loop rewrite only the code around it. That code layer is the harness, the runtime wrapper that feeds the model context, runs its tool calls, and decides when a run ends. By the time the loop finished, the system had essentially matched Sonnet 4.6 on the benchmark's headline metric, at roughly 7x lower cost per task. Zero weights changed. The gain existed because of where the model was failing. The judge only grades files saved in the right place under the exact requested filename, and the model kept doing the legal analysis correctly, then saving it under the wrong name, dropping it in a scratch folder, or never writing it at all. So the 0% was never measuring legal reasoning. It was measuring the harness. Hand-tuning that layer is slow and model-specific, so they automated it. A Claude proposer adds exactly one mechanism per iteration, and an outer loop keeps it only if it clearly beats the current best, so accepted mechanisms compound. What the loop discovered says a lot about where agents actually fail. → The biggest single gain was file handling, not intelligence. An automatic step that lands the deliverable exactly where the judge expects it beat every prompt change, with zero extra model tokens. → Code fixes transferred across models, prompt playbooks did not. The same harness lifted a smaller model from the same family by 14 points, but the tuned prompts hurt a different model family on tasks it could already finish. → The harness mattered more than anything else. Same model, same judge, same tasks, and five different harnesses scored anywhere between 3.5% and 80.1%. The gains do eventually flatten, and the remaining misses look like real capability gaps. At some point the wrapper runs out of tricks and the model has to carry the work. But the lesson holds. A benchmark score measures the model and its harness together, and until the harness is fixed, it's impossible to know which one failed. I highly recommend reading this: I also wrote a deep dive on agent harness engineering a while back, covering the orchestration loop, tools, memory, context management, and everything that turns a stateless LLM into a capable agent. The article is quoted below.
0:18

Don't train the model, evolve the harness. I read a brilliant blog post from Hugging Face where they took a frozen open model scoring 0% on a hard legal agent benchmark, left its weights alone, and let an automated loop rewrite only the code around it. That code layer is the harness, the runtime wrapper that feeds the model context, runs its tool calls, and decides when a run ends. By the time the loop finished, the system had essentially matched Sonnet 4.6 on the benchmark's headline metric, at roughly 7x lower cost per task. Zero weights changed. The gain existed because of where the model was failing. The judge only grades files saved in the right place under the exact requested filename, and the model kept doing the legal analysis correctly, then saving it under the wrong name, dropping it in a scratch folder, or never writing it at all. So the 0% was never measuring legal reasoning. It was measuring the harness. Hand-tuning that layer is slow and model-specific, so they automated it. A Claude proposer adds exactly one mechanism per iteration, and an outer loop keeps it only if it clearly beats the current best, so accepted mechanisms compound. What the loop discovered says a lot about where agents actually fail. → The biggest single gain was file handling, not intelligence. An automatic step that lands the deliverable exactly where the judge expects it beat every prompt change, with zero extra model tokens. → Code fixes transferred across models, prompt playbooks did not. The same harness lifted a smaller model from the same family by 14 points, but the tuned prompts hurt a different model family on tasks it could already finish. → The harness mattered more than anything else. Same model, same judge, same tasks, and five different harnesses scored anywhere between 3.5% and 80.1%. The gains do eventually flatten, and the remaining misses look like real capability gaps. At some point the wrapper runs out of tricks and the model has to carry the work. But the lesson holds. A benchmark score measures the model and its harness together, and until the harness is fixed, it's impossible to know which one failed. I highly recommend reading this: I also wrote a deep dive on agent harness engineering a while back, covering the orchestration loop, tools, memory, context management, and everything that turns a stateless LLM into a capable agent. The article is quoted below.

Akshay 🚀

244,402 views • 1 month ago

We released physics-intern: a simple harness for science problems! It gets models like Gemini 3.1 Pro to go from 17.7 -> 31.4, thus beating GPT 5.5 Pro. The physics-intern harness can wrap any model and via dedicated subagent boost the performance of the vanilla reasoning models. While I think more and more of these harness capability gains will be absorbed into the models (like prompting tricks disappeared over time) there is a lot to be gained right now by building good scaffolds for those models and integrating tools well. Interestingly, the exception we found that GPT 5.5 Pro actually didn't benefit from the physics-intern harness! Read more about it here: PS: I think the Harness[Model] notation is kind of nice.
0:12

We released physics-intern: a simple harness for science problems! It gets models like Gemini 3.1 Pro to go from 17.7 -> 31.4, thus beating GPT 5.5 Pro. The physics-intern harness can wrap any model and via dedicated subagent boost the performance of the vanilla reasoning models. While I think more and more of these harness capability gains will be absorbed into the models (like prompting tricks disappeared over time) there is a lot to be gained right now by building good scaffolds for those models and integrating tools well. Interestingly, the exception we found that GPT 5.5 Pro actually didn't benefit from the physics-intern harness! Read more about it here: PS: I think the Harness[Model] notation is kind of nice.

Leandro von Werra

97,262 views • 2 months ago

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. Akshay wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. Read it below.
0:15

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. Akshay wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. Read it below.

Avi Chawla

439,852 views • 29 days ago

you can run claude code inside antigravity completely Free with zero credit card and no rate limits 😳 use openrouter’s free models + antigravity. no anthropic bill. no paid api keys. takes 10 minutes to set up. what you get during this setup: - full claude code agent experience - strong coding models (including deepseek-r1, qwen2.5-coder, llama-4, grok-4 free tier) - antigravity’s clean workspace and sandbox - unlimited usage (as long as you stay on free models) - easy model swapping - zero cost full setup guide (100% free): step 1: install antigravity -go to and install it -create a new workspace step 2: install claude code - inside antigravity, install the claude code extension from the marketplace - open the built-in terminal step 3: create openrouter free account -go to - sign up with google (no card needed) - go to keys and create a new api key step 4: set the environment variables -in antigravity terminal run: export ANTHROPIC_API_KEY=sk-or-xxx export OPENROUTER_API_KEY=sk-or-xxx step 5: launch claude code with free model -run this command: claude-code --model deepseek/deepseek-r1:free or try: qwen/qwen2.5-coder:free if you already have antigravity? skip straight to step 2. after 10 minutes you’ll have a full agentic coding setup running for free. this is currently one of the cheapest ways to run serious coding agents in 2026. bookmark this before they limit the free models.
0:17

you can run claude code inside antigravity completely Free with zero credit card and no rate limits 😳 use openrouter’s free models + antigravity. no anthropic bill. no paid api keys. takes 10 minutes to set up. what you get during this setup: - full claude code agent experience - strong coding models (including deepseek-r1, qwen2.5-coder, llama-4, grok-4 free tier) - antigravity’s clean workspace and sandbox - unlimited usage (as long as you stay on free models) - easy model swapping - zero cost full setup guide (100% free): step 1: install antigravity -go to and install it -create a new workspace step 2: install claude code - inside antigravity, install the claude code extension from the marketplace - open the built-in terminal step 3: create openrouter free account -go to - sign up with google (no card needed) - go to keys and create a new api key step 4: set the environment variables -in antigravity terminal run: export ANTHROPIC_API_KEY=sk-or-xxx export OPENROUTER_API_KEY=sk-or-xxx step 5: launch claude code with free model -run this command: claude-code --model deepseek/deepseek-r1:free or try: qwen/qwen2.5-coder:free if you already have antigravity? skip straight to step 2. after 10 minutes you’ll have a full agentic coding setup running for free. this is currently one of the cheapest ways to run serious coding agents in 2026. bookmark this before they limit the free models.

painn

32,057 views • 1 month ago

I explored a further possibility with local models: Qwen3.6 35B A3B + NVIDIA LocateAnything-3B as a local Computer Use agent (proof of concept). In the demo, I asked it to switch my Mac to light mode. It did. Then back to dark. Did that too — finding the right toggle in System Settings, clicking it, and verifying the change itself. It's fully screenshot-based, so no Accessibility API needed. If it's on screen, the agent can see it and act on it. This runs entirely on your own hardware — private, local, built from two small open models.
0:16

I explored a further possibility with local models: Qwen3.6 35B A3B + NVIDIA LocateAnything-3B as a local Computer Use agent (proof of concept). In the demo, I asked it to switch my Mac to light mode. It did. Then back to dark. Did that too — finding the right toggle in System Settings, clicking it, and verifying the change itself. It's fully screenshot-based, so no Accessibility API needed. If it's on screen, the agent can see it and act on it. This runs entirely on your own hardware — private, local, built from two small open models.

stevibe

44,052 views • 2 months ago

✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!
0:51

✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!

@levelsio

119,210 views • 1 year ago

50% cheaper Claude inference with just one line of code change! - Remove → model="claude-opus-4-8" - Add → model="ship-like/claude-opus-4-8" I verified the cost saving in my own terminal by invoking the same Anthropic model with the same prompt. The underlying engineering by Ship is actually interesting, and the patterns can be used in any production LLM stack. Essentially, a trained model is a frozen artifact. Every request performs the same forward-pass, whether it extracts a date or refactors a module, because the compute decision was made at training time, before the request existed. Ship makes that decision at inference time instead. After seeing a request, it searches over executions, involving single models, cascades, ensembles, or harnesses with tools, and serves the cheapest one that will match the reference model's quality. This is not a basic router, because picking a cheaper model per query doesn't ensure the cheaper model preserves the original's behavior, like output shape, tool-call patterns, and refusals. Ship measures this equivalence directly. Outputs stay distributionally indistinguishable from the reference model, not token-identical, since two calls to the same model already differ, but they are indistinguishable in capability and behavior. Of course, some requests execute cheaply and some cost Ship more than the customer pays, but the price per request is still a flat 50% off either way, so the execution-cost variance moves off the application's bill entirely. The video below depicts the cost savings and output in my real invocation, and I partnered with the team to put this together.
0:12

50% cheaper Claude inference with just one line of code change! - Remove → model="claude-opus-4-8" - Add → model="ship-like/claude-opus-4-8" I verified the cost saving in my own terminal by invoking the same Anthropic model with the same prompt. The underlying engineering by Ship is actually interesting, and the patterns can be used in any production LLM stack. Essentially, a trained model is a frozen artifact. Every request performs the same forward-pass, whether it extracts a date or refactors a module, because the compute decision was made at training time, before the request existed. Ship makes that decision at inference time instead. After seeing a request, it searches over executions, involving single models, cascades, ensembles, or harnesses with tools, and serves the cheapest one that will match the reference model's quality. This is not a basic router, because picking a cheaper model per query doesn't ensure the cheaper model preserves the original's behavior, like output shape, tool-call patterns, and refusals. Ship measures this equivalence directly. Outputs stay distributionally indistinguishable from the reference model, not token-identical, since two calls to the same model already differ, but they are indistinguishable in capability and behavior. Of course, some requests execute cheaply and some cost Ship more than the customer pays, but the price per request is still a flat 50% off either way, so the execution-cost variance moves off the application's bill entirely. The video below depicts the cost savings and output in my real invocation, and I partnered with the team to put this together.

Akshay 🚀

63,725 views • 13 days ago

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 👇
0:15

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 👇

Ziwen

93,988 views • 3 days ago

90% of "AI developers" just download pre packaged GGUF files from Hugging Face, hit run, and call it a day. The top 10% know how to pull the raw safetensors, run the math, and quantize massive models into Q4_K_M themselves. If you think llama.cpp can only execute models, you’re missing the best part of the open source ecosystem. It’s a high performance optimization suite. Manually stripping 69% of the VRAM footprint off a brand new model architecture is where real infrastructure value is made. If you want to actually master local inference and deploy models like Google’s massive Gemma 4 12B it on consumer NVIDIA hardware using llama.cpp, you need to learn this pipeline. Let's build it. I just took the raw 22.7 GB Gemma 4 baseline and manually compressed it down to a 7.02 GB Q4_K_M GGUF artifact using llama.cpp. That is a 69% reduction in footprint. No quality loss. No VRAM bottlenecks. Just native, hardware accelerated C++ inference running a full 2,50,000 token context window on a dual NVIDIA Tesla T4 setup. Stop melting your VRAM on unoptimized weights and stop relying on other people's pipelines. Own your stack. I mapped this entire architecture from dynamic binary fetching to raw quantization and real time GPU streaming into a single, bulletproof notebook. Notebook link is in the comments below. Bookmark this blueprint for your next deployment and tell me which quantization works best for your workflow and model.
0:19

90% of "AI developers" just download pre packaged GGUF files from Hugging Face, hit run, and call it a day. The top 10% know how to pull the raw safetensors, run the math, and quantize massive models into Q4_K_M themselves. If you think llama.cpp can only execute models, you’re missing the best part of the open source ecosystem. It’s a high performance optimization suite. Manually stripping 69% of the VRAM footprint off a brand new model architecture is where real infrastructure value is made. If you want to actually master local inference and deploy models like Google’s massive Gemma 4 12B it on consumer NVIDIA hardware using llama.cpp, you need to learn this pipeline. Let's build it. I just took the raw 22.7 GB Gemma 4 baseline and manually compressed it down to a 7.02 GB Q4_K_M GGUF artifact using llama.cpp. That is a 69% reduction in footprint. No quality loss. No VRAM bottlenecks. Just native, hardware accelerated C++ inference running a full 2,50,000 token context window on a dual NVIDIA Tesla T4 setup. Stop melting your VRAM on unoptimized weights and stop relying on other people's pipelines. Own your stack. I mapped this entire architecture from dynamic binary fetching to raw quantization and real time GPU streaming into a single, bulletproof notebook. Notebook link is in the comments below. Bookmark this blueprint for your next deployment and tell me which quantization works best for your workflow and model.

Alok

62,631 views • 24 days ago

DeepSeek R1 is *the* best model available right now. It's at the level of o1, but you can use it for free, and it's much faster. A huge leap forward that nobody saw coming. No wonder so many people are throwing tantrums online trying to discredit the Chinese students who built this. You can use DeepSeek in Visual Studio Code right now: 1. Install the Qodo Gen AI extension 2. Select DeepSeek R1 from their list of models The Qodo team is hosting DeepSeek on their servers, so none of your data will go to China. I've been building a Tetris game using DeepSeek, and this is the most impressive model I've seen so far.
0:17

DeepSeek R1 is *the* best model available right now. It's at the level of o1, but you can use it for free, and it's much faster. A huge leap forward that nobody saw coming. No wonder so many people are throwing tantrums online trying to discredit the Chinese students who built this. You can use DeepSeek in Visual Studio Code right now: 1. Install the Qodo Gen AI extension 2. Select DeepSeek R1 from their list of models The Qodo team is hosting DeepSeek on their servers, so none of your data will go to China. I've been building a Tetris game using DeepSeek, and this is the most impressive model I've seen so far.

Santiago

1,224,151 views • 1 year ago

introducing a new, very fun, LLM benchmark- the Game-of-Life Bench! the rules are simple: given an 8x8 grid following Conway's game of life rules, the goal is to create an initial pattern with at most 32 cells that can last the longest number of turns before dying/repeating. some results to highlight (with caveats detailed below): - gpt 5.1 lasts the longest with a 106 step run - claude models are really bad at this! they refuse to reason about this task and score < 25 points - deepseek r1 is the best open model with 102 steps. why? because i wanted to create a benchmark that has (i think) no practicality, but is still fun to look at, cheap, and still measures something interesting. i also am a big fan of the game of life. its absurdly simple rules leading to intractability is extremely cool to me. also, i saw a lot of work with LLMs trying to "predict" the next state in Conway's game of life, I think game-of-life bench is more fun because it's pretty open ended and only asks the LLM for the initial state. I also think this could be an RL env? but idk why you would ever train on this task haha i don't think this is a "serious" benchmark because it doesnt measure anything practical, but i still think it's a hard benchmark exactly because you can't predict what happens with your initial state many turns into the future; this is why i was initially expecting all LLMs to be bad at it, but turns out, some are clearly better than the others (the ordering may surprise you!) reminder: this is still a work-in-progress; (1) i am gpu-poor so could only do 10 runs for each model, even though total running cost is relatively low. maybe with some more credits i can run more seeds for each model. (2) i handpicked models which i think are at the frontier right now, plus some others that were on my mind. so, if you'd like to see a model on here, let me know. (3) i currently only do an 8x8 grid because i thought that by itself would be pretty hard for current LLMs, but of course we can increase grid sizes! (4) the coolest thing is, i dont think we can calculate the max possible number of states (yay undecidability!) you can go without repeating, so this is essentially a no-ceiling task, which is pretty cool! again, i did this mostly out of a desire to make LLMs do something fun. if this keeps me entertained for a few more days, i'd likely release a blog post on it. if it keeps me entertained for a week (and someone sponsors me), i'll put more work into it :P lastly, this is fully open sourced, so feel free to run this on your own!
0:19

introducing a new, very fun, LLM benchmark- the Game-of-Life Bench! the rules are simple: given an 8x8 grid following Conway's game of life rules, the goal is to create an initial pattern with at most 32 cells that can last the longest number of turns before dying/repeating. some results to highlight (with caveats detailed below): - gpt 5.1 lasts the longest with a 106 step run - claude models are really bad at this! they refuse to reason about this task and score < 25 points - deepseek r1 is the best open model with 102 steps. why? because i wanted to create a benchmark that has (i think) no practicality, but is still fun to look at, cheap, and still measures something interesting. i also am a big fan of the game of life. its absurdly simple rules leading to intractability is extremely cool to me. also, i saw a lot of work with LLMs trying to "predict" the next state in Conway's game of life, I think game-of-life bench is more fun because it's pretty open ended and only asks the LLM for the initial state. I also think this could be an RL env? but idk why you would ever train on this task haha i don't think this is a "serious" benchmark because it doesnt measure anything practical, but i still think it's a hard benchmark exactly because you can't predict what happens with your initial state many turns into the future; this is why i was initially expecting all LLMs to be bad at it, but turns out, some are clearly better than the others (the ordering may surprise you!) reminder: this is still a work-in-progress; (1) i am gpu-poor so could only do 10 runs for each model, even though total running cost is relatively low. maybe with some more credits i can run more seeds for each model. (2) i handpicked models which i think are at the frontier right now, plus some others that were on my mind. so, if you'd like to see a model on here, let me know. (3) i currently only do an 8x8 grid because i thought that by itself would be pretty hard for current LLMs, but of course we can increase grid sizes! (4) the coolest thing is, i dont think we can calculate the max possible number of states (yay undecidability!) you can go without repeating, so this is essentially a no-ceiling task, which is pretty cool! again, i did this mostly out of a desire to make LLMs do something fun. if this keeps me entertained for a few more days, i'd likely release a blog post on it. if it keeps me entertained for a week (and someone sponsors me), i'll put more work into it :P lastly, this is fully open sourced, so feel free to run this on your own!

Akshit

13,722 views • 5 months ago

On your open perpetual position can become an NFT and live in your wallet. You open a position like normal than mint it. The program issues a token 2022 nft, that points to your exact position. Normally your open position lives inside the protocol. With Percolator Trade your able to wrap it into an NFT that sits in your own wallet. A real, ownable object instead of a line in a dashboard. Because it’s a token, it moves like one. Send the NFT to another wallet and the position goes with it. Whoever holds the NFT owns the trade. One NFT = one position If you burn it, your back to holding the position directly. That’s the composable part, a position use to be stuck inside one account. Because it’s a token, it can now be sold OTC or used by anything that speaks tokens. Why does it matter? Because the position is bound to the NFT, it can’t be drained out from under whoever holds it so there is no middleman that holds your position. The NFT does, and the NFT is yours. One important safety detail; transfers run through a token 2022 hook that checks the positions health first, so it can’t be handed off in a broken state & the NFT layer is a separate program so it can never touch the funds the core engine holds.
0:13

On your open perpetual position can become an NFT and live in your wallet. You open a position like normal than mint it. The program issues a token 2022 nft, that points to your exact position. Normally your open position lives inside the protocol. With Percolator Trade your able to wrap it into an NFT that sits in your own wallet. A real, ownable object instead of a line in a dashboard. Because it’s a token, it moves like one. Send the NFT to another wallet and the position goes with it. Whoever holds the NFT owns the trade. One NFT = one position If you burn it, your back to holding the position directly. That’s the composable part, a position use to be stuck inside one account. Because it’s a token, it can now be sold OTC or used by anything that speaks tokens. Why does it matter? Because the position is bound to the NFT, it can’t be drained out from under whoever holds it so there is no middleman that holds your position. The NFT does, and the NFT is yours. One important safety detail; transfers run through a token 2022 hook that checks the positions health first, so it can’t be handed off in a broken state & the NFT layer is a separate program so it can never touch the funds the core engine holds.

Percolator

12,359 views • 1 month ago

SOMEONE BUILT A TERMINAL TORRENT CLIENT THAT SEARCHES EVERY TRUSTED SOURCE AT ONCE AND DOWNLOADS STRAIGHT TO DISK finding a torrent in 2026 is miserable fake download buttons, popups that spawn three more tabs, and half the results are dead with zero seeders. so he built the opposite of all that: > type one query and it hits a curated set of trackers all at once > results stream back tagged with source, size and seeders as fast as each site answers > arrow to the one you want, press d, and it lands on your drive > if one source is down it just skips it and keeps searching the rest > downloads run in the background and pick up where they left off if you quit mid transfer the whole thing is one command. npx and youre running, all you need is node, no browser, no setup, nothing leaves your machine except the request itself. its called torlink, open source and mit. this is the kind of clean terminal tool that does one annoying thing perfectly
0:16

SOMEONE BUILT A TERMINAL TORRENT CLIENT THAT SEARCHES EVERY TRUSTED SOURCE AT ONCE AND DOWNLOADS STRAIGHT TO DISK finding a torrent in 2026 is miserable fake download buttons, popups that spawn three more tabs, and half the results are dead with zero seeders. so he built the opposite of all that: > type one query and it hits a curated set of trackers all at once > results stream back tagged with source, size and seeders as fast as each site answers > arrow to the one you want, press d, and it lands on your drive > if one source is down it just skips it and keeps searching the rest > downloads run in the background and pick up where they left off if you quit mid transfer the whole thing is one command. npx and youre running, all you need is node, no browser, no setup, nothing leaves your machine except the request itself. its called torlink, open source and mit. this is the kind of clean terminal tool that does one annoying thing perfectly

Om Patel

359,068 views • 1 month ago

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.
0:17

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.

Akshay 🚀

31,249 views • 6 months ago

Why is the market selling off today? (Save this). The semi selloff right now is being driven by a mix of macro fear, profit taking and investors questioning how quickly all of this AI spending will actually pay off, not because demand for AI infrastructure suddenly disappeared. The market is basically trading this chain reaction, the ongoing US Iran escalation pushes oil higher, higher oil keeps inflation elevated, sticky inflation keeps Treasury yields high and that increases the risk of the Fed staying hawkish or even hiking again. That is a terrible setup for semis because many of these companies are valued on the massive earnings investors expect them to generate years from now. When yields rise, those future earnings become worth less today which is why the highest multiple AI and semiconductor names usually get hit first. (I don't think there will be a hike this year). This is also why everything is moving together right now. Nvidia, Micron, Nebius, SanDisk, Broadcom and Applied Optoelectronics are all completely different businesses, but institutions are not separating memory, networking, optics, compute and cloud infrastructure at the moment. They are reducing exposure to the entire AI trade, taking profits in the names that have already run the most and moving into a more defensive position potentially ahead of the Fed. There is also growing pressure around hyperscaler capex. Microsoft, Meta, Amazon and Google are still spending enormous amounts on GPUs, data centers, networking and power but the market is starting to ask when all of that spending will actually turn into revenue and free cash flow. Investors are no longer satisfied with hearing that AI capex is growing. They want proof that the returns are arriving fast enough to justify the valuations already priced into the entire AI ecosystem. That creates a weird situation where hyperscaler capex can continue rising while semiconductor stocks still fall. The market is not asking whether AI spending is growing anymore but rather asking whether it is growing fast enough to beat the expectations already baked into these stocks. Crowded positioning is another major factor. Semis and AI infrastructure stocks have been some of the biggest winners in the market so institutions are sitting on huge profits and many funds own the exact same names. When macro risk increases, investors usually sell the most liquid winners first. That does not mean demand for memory, optics or custom chips suddenly collapsed but rather means investors are locking in gains and reducing risk. Tariffs add another layer because even when they are not directly placed on chips, they can still raise the cost of servers, electrical equipment, cooling systems, construction materials and the overall data center buildout. That makes AI infrastructure more expensive while also adding another source of inflation. Then you have Jensen Huang’s letter to the White House this morning about open weight AI models, which I think is one of the most important long term developments here. Nvidia, Meta, Microsoft, Palantir and several other companies are pushing Washington not to place broad restrictions on open weight AI. OpenAI and Anthropic were notably absent because open models are much more of a threat to their business models. OpenAI and Anthropic benefit from a world where a few closed frontier labs control the best models and companies have to pay them through subscriptions and APIs. Open weight models weaken that advantage because businesses can download a model, customize it for their own use and run it on their own infrastructure or through a neocloud. That is bad for OpenAI and Anthropic because it puts pressure on pricing, margins and the idea that they will control the intelligence layer of the economy but it is very good for the AI ecosystem as a whole over the long run. But the question is what does this mean for all the OpenAI and Anthropic commitments? so that's adding to the fear as well. But with that being said open models make AI cheaper and more accessible. Instead of AI being controlled by a few giant labs, thousands of startups, universities, governments and regular businesses can deploy models themselves. That spreads AI adoption across the entire economy and creates a much larger infrastructure opportunity and that is exactly why Jensen cares. Nvidia does not need OpenAI or Anthropic to win. Nvidia just needs more people using AI. Whether the model comes from OpenAI, Anthropic, Meta, Mistral, Kimi or some startup nobody has heard of yet, it still needs GPUs, memory, networking, data centers and electricity. So open weight AI could actually weaken the model companies while making the infrastructure layer much bigger. More open models mean more companies running inference. More inference means more GPUs. More GPUs mean more HBM, optical transceivers, switches, data centers and power. That is bullish for Nvidia Nebius, Micron, Broadcom , Marvell and Applied Optoelectronics over the long run. So my take is that the current semi selloff is being driven mostly by macro uncertainty, higher oil, rising yields, Fed fears, tariffs, crowded positioning and questions around the return on hyperscaler capex. The underlying AI infrastructure thesis has not suddenly broken. We are not broadly seeing hyperscalers cancel GPU orders, slash capex, abandon data center projects or report that AI demand has collapsed. What has changed is the valuation investors are willing to pay while the macro environment remains unstable. The market is lowering the price it is willing to pay for semiconductor growth but is not necessarily saying that growth is gone. And while Jensen’s open weight push may be bad for OpenAI and Anthropic, it could be one of the best things possible for the AI ecosystem over the long run because it creates more models, more developers, more competition and ultimately much more demand for the infrastructure underneath all of it. Nothing about the AI thesis has changed for me, so I will be going shopping and taking advantage of this sale while the market is selling everything together. I am an analyst at Milk Road Pro, and if you want to see exactly what I am buying, you can join for just $1 using the link below.
0:11

Why is the market selling off today? (Save this). The semi selloff right now is being driven by a mix of macro fear, profit taking and investors questioning how quickly all of this AI spending will actually pay off, not because demand for AI infrastructure suddenly disappeared. The market is basically trading this chain reaction, the ongoing US Iran escalation pushes oil higher, higher oil keeps inflation elevated, sticky inflation keeps Treasury yields high and that increases the risk of the Fed staying hawkish or even hiking again. That is a terrible setup for semis because many of these companies are valued on the massive earnings investors expect them to generate years from now. When yields rise, those future earnings become worth less today which is why the highest multiple AI and semiconductor names usually get hit first. (I don't think there will be a hike this year). This is also why everything is moving together right now. Nvidia, Micron, Nebius, SanDisk, Broadcom and Applied Optoelectronics are all completely different businesses, but institutions are not separating memory, networking, optics, compute and cloud infrastructure at the moment. They are reducing exposure to the entire AI trade, taking profits in the names that have already run the most and moving into a more defensive position potentially ahead of the Fed. There is also growing pressure around hyperscaler capex. Microsoft, Meta, Amazon and Google are still spending enormous amounts on GPUs, data centers, networking and power but the market is starting to ask when all of that spending will actually turn into revenue and free cash flow. Investors are no longer satisfied with hearing that AI capex is growing. They want proof that the returns are arriving fast enough to justify the valuations already priced into the entire AI ecosystem. That creates a weird situation where hyperscaler capex can continue rising while semiconductor stocks still fall. The market is not asking whether AI spending is growing anymore but rather asking whether it is growing fast enough to beat the expectations already baked into these stocks. Crowded positioning is another major factor. Semis and AI infrastructure stocks have been some of the biggest winners in the market so institutions are sitting on huge profits and many funds own the exact same names. When macro risk increases, investors usually sell the most liquid winners first. That does not mean demand for memory, optics or custom chips suddenly collapsed but rather means investors are locking in gains and reducing risk. Tariffs add another layer because even when they are not directly placed on chips, they can still raise the cost of servers, electrical equipment, cooling systems, construction materials and the overall data center buildout. That makes AI infrastructure more expensive while also adding another source of inflation. Then you have Jensen Huang’s letter to the White House this morning about open weight AI models, which I think is one of the most important long term developments here. Nvidia, Meta, Microsoft, Palantir and several other companies are pushing Washington not to place broad restrictions on open weight AI. OpenAI and Anthropic were notably absent because open models are much more of a threat to their business models. OpenAI and Anthropic benefit from a world where a few closed frontier labs control the best models and companies have to pay them through subscriptions and APIs. Open weight models weaken that advantage because businesses can download a model, customize it for their own use and run it on their own infrastructure or through a neocloud. That is bad for OpenAI and Anthropic because it puts pressure on pricing, margins and the idea that they will control the intelligence layer of the economy but it is very good for the AI ecosystem as a whole over the long run. But the question is what does this mean for all the OpenAI and Anthropic commitments? so that's adding to the fear as well. But with that being said open models make AI cheaper and more accessible. Instead of AI being controlled by a few giant labs, thousands of startups, universities, governments and regular businesses can deploy models themselves. That spreads AI adoption across the entire economy and creates a much larger infrastructure opportunity and that is exactly why Jensen cares. Nvidia does not need OpenAI or Anthropic to win. Nvidia just needs more people using AI. Whether the model comes from OpenAI, Anthropic, Meta, Mistral, Kimi or some startup nobody has heard of yet, it still needs GPUs, memory, networking, data centers and electricity. So open weight AI could actually weaken the model companies while making the infrastructure layer much bigger. More open models mean more companies running inference. More inference means more GPUs. More GPUs mean more HBM, optical transceivers, switches, data centers and power. That is bullish for Nvidia Nebius, Micron, Broadcom , Marvell and Applied Optoelectronics over the long run. So my take is that the current semi selloff is being driven mostly by macro uncertainty, higher oil, rising yields, Fed fears, tariffs, crowded positioning and questions around the return on hyperscaler capex. The underlying AI infrastructure thesis has not suddenly broken. We are not broadly seeing hyperscalers cancel GPU orders, slash capex, abandon data center projects or report that AI demand has collapsed. What has changed is the valuation investors are willing to pay while the macro environment remains unstable. The market is lowering the price it is willing to pay for semiconductor growth but is not necessarily saying that growth is gone. And while Jensen’s open weight push may be bad for OpenAI and Anthropic, it could be one of the best things possible for the AI ecosystem over the long run because it creates more models, more developers, more competition and ultimately much more demand for the infrastructure underneath all of it. Nothing about the AI thesis has changed for me, so I will be going shopping and taking advantage of this sale while the market is selling everything together. I am an analyst at Milk Road Pro, and if you want to see exactly what I am buying, you can join for just $1 using the link below.

Melvin

179,241 views • 10 days ago

i built an open world minigame that's controlled with hand movements only here's a step-by-step tutorial: > started with my mediapipe + threejs template (see QT) > added a 3D model made by quaternius > sent a few prompts to gemini 3... --- prompt #1: repurpose the attached script, but now I want to import a gltf model (assets/model.gltf) and use that in the scene instead of the cube when the user moves their hand, a waypoint indicator should move around the scene, and the 3D model should smoothly move there when the user makes a fist, the model should jump the 3D model contains bundled animations in it. use the "idle", "run", and "jump" animations --- prompt #2: make this a procedural open world adventure. generate very simple procedural voxel terrain when the model gets near the edges of the screen, the camera should move, allowing the model to keep going in that direction --- prompt #3: add some floating glowing gems around the map that the user can collect --- i made some manual tweaks for styling and game feel, but that's the gist of it thanks for reading if you got all the way here full code is available at my link in bio
0:15

i built an open world minigame that's controlled with hand movements only here's a step-by-step tutorial: > started with my mediapipe + threejs template (see QT) > added a 3D model made by quaternius > sent a few prompts to gemini 3... --- prompt #1: repurpose the attached script, but now I want to import a gltf model (assets/model.gltf) and use that in the scene instead of the cube when the user moves their hand, a waypoint indicator should move around the scene, and the 3D model should smoothly move there when the user makes a fist, the model should jump the 3D model contains bundled animations in it. use the "idle", "run", and "jump" animations --- prompt #2: make this a procedural open world adventure. generate very simple procedural voxel terrain when the model gets near the edges of the screen, the camera should move, allowing the model to keep going in that direction --- prompt #3: add some floating glowing gems around the map that the user can collect --- i made some manual tweaks for styling and game feel, but that's the gist of it thanks for reading if you got all the way here full code is available at my link in bio

AA

131,799 views • 8 months ago

Moonshot AI is casually giving developers free daily access to Kimi K3 😳 no subscription no upfront payment just sign in and start using one of the largest open AI models available what you get for $0: - Kimi K3 with 2.8T parameters - 1M token context window - strong coding and reasoning performance - native vision capabilities - free daily credits that refresh automatically why this is worth checking: > access a frontier model without paying API fees > long context for large codebases and documents > works on web, desktop, mobile, and CLI getting started takes less than 2 minutes: 1. go to 2. create a free account 3. Kimi K3 is available as the default model 4. start chatting or coding with your daily free credits bonus: Moonshot Together lets you invite friends for a chance to earn 3, 7, 15, 30, or even 365 days of Kimi Membership through its rewards program benchmark highlights: > 2.8T parameter MoE model > 1M context window > strong performance across coding, browsing, and reasoning benchmarks important: free credits reset daily, rate limits apply on the free tier, and the open-weight release is expected on July 27 A simple way to try one of the latest frontier AI models without paying for API access
0:17

Moonshot AI is casually giving developers free daily access to Kimi K3 😳 no subscription no upfront payment just sign in and start using one of the largest open AI models available what you get for $0: - Kimi K3 with 2.8T parameters - 1M token context window - strong coding and reasoning performance - native vision capabilities - free daily credits that refresh automatically why this is worth checking: > access a frontier model without paying API fees > long context for large codebases and documents > works on web, desktop, mobile, and CLI getting started takes less than 2 minutes: 1. go to 2. create a free account 3. Kimi K3 is available as the default model 4. start chatting or coding with your daily free credits bonus: Moonshot Together lets you invite friends for a chance to earn 3, 7, 15, 30, or even 365 days of Kimi Membership through its rewards program benchmark highlights: > 2.8T parameter MoE model > 1M context window > strong performance across coding, browsing, and reasoning benchmarks important: free credits reset daily, rate limits apply on the free tier, and the open-weight release is expected on July 27 A simple way to try one of the latest frontier AI models without paying for API access

K2S

22,890 views • 16 days ago

The architecture of this new world model is one of the most interesting things I've seen lately: Let me first explain how most world models work: They predict and render one frame at a time. If you are navigating in one of these worlds, and you look left, the model draws whatever looks right in the moment. Every time you change your viewpoint, the model has to imagine what should be there again, so it's very common for these models to "forget" what's in the world. For example, if you put a toy on the table, look away, then look back, the toy might not be there anymore. Tripo AI is releasing its Project Eden model, which works very differently: The model builds the world first, and then renders it based on that map. That map holds the real state of the world: the geometry, every object, where things are, what's already happened. The picture you see on screen gets generated from the map. This architecture flips the whole thing. Now, you get the following: 1. The world stops forgetting. Leave, come back, and the toy is still on the table because it lives in the map, not in the last frame you saw. 2. You can edit the world, and those changes persist for anyone who enters later. 3. Multiple people and AI agents can coexist in the world and see it from different perspectives. This is early research, but it's looking really promising. They just raised nearly $200M across two rounds to build it out. Tripo will be at SIGGRAPH 2026 (July 19–23, Los Angeles Convention Center). If you work in 3D, embodied AI, simulation, or anything spatial, go connect with them there.
0:10

The architecture of this new world model is one of the most interesting things I've seen lately: Let me first explain how most world models work: They predict and render one frame at a time. If you are navigating in one of these worlds, and you look left, the model draws whatever looks right in the moment. Every time you change your viewpoint, the model has to imagine what should be there again, so it's very common for these models to "forget" what's in the world. For example, if you put a toy on the table, look away, then look back, the toy might not be there anymore. Tripo AI is releasing its Project Eden model, which works very differently: The model builds the world first, and then renders it based on that map. That map holds the real state of the world: the geometry, every object, where things are, what's already happened. The picture you see on screen gets generated from the map. This architecture flips the whole thing. Now, you get the following: 1. The world stops forgetting. Leave, come back, and the toy is still on the table because it lives in the map, not in the last frame you saw. 2. You can edit the world, and those changes persist for anyone who enters later. 3. Multiple people and AI agents can coexist in the world and see it from different perspectives. This is early research, but it's looking really promising. They just raised nearly $200M across two rounds to build it out. Tripo will be at SIGGRAPH 2026 (July 19–23, Los Angeles Convention Center). If you work in 3D, embodied AI, simulation, or anything spatial, go connect with them there.

Santiago

30,189 views • 1 month ago

Fable 5 comes back！It can now build playable game prototypes. I think it is actually a signal for where AI coding is going. Making a game is not just “write some code.” Even a small browser game needs: game loop；character movement；collision logic；scoring system；UI states；physics tuning；visual feedback；bug fixing；playtesting This is why game prototyping is a great test for AI models. A model cannot fake it with a pretty answer. Either the game runs, or it does not. What impressed me about Fable 5 is that it is useful for the messy middle: turning an idea into mechanics, turning mechanics into code, debugging broken interactions, and iterating until the prototype feels playable. But here is the practical part: I would not use the strongest model for every step. For game building, I would split the workflow: 1. Fable 5 for game design + architecture 2. a fast coding model for routine implementation 3. a vision-capable model for screenshot/UI feedback 4. a cheaper model for docs, test cases, and small fixes 5. fallback when latency, cost, or output quality becomes a problem That is the real AI coding stack. Not “one magic model does everything.” More like: the right model, for the right task, at the right cost, with fallback when things break. This is why I’ve been looking at ZenMux ZenMux. ZenMux gives developers one gateway to access multiple leading AI models, with OpenAI / Anthropic / Google Vertex compatible APIs, cost tracking, quality benchmarks, auto-routing, and compensation when output quality, latency, or throughput falls short. If AI can now make games, the next question is not just “which model is strongest?” It is:how do we manage the whole model workflow Fable 5 shows the creative ceiling. ZenMux is closer to the infrastructure layer you need when AI coding becomes a real production habit.
0:12

Fable 5 comes back！It can now build playable game prototypes. I think it is actually a signal for where AI coding is going. Making a game is not just “write some code.” Even a small browser game needs: game loop；character movement；collision logic；scoring system；UI states；physics tuning；visual feedback；bug fixing；playtesting This is why game prototyping is a great test for AI models. A model cannot fake it with a pretty answer. Either the game runs, or it does not. What impressed me about Fable 5 is that it is useful for the messy middle: turning an idea into mechanics, turning mechanics into code, debugging broken interactions, and iterating until the prototype feels playable. But here is the practical part: I would not use the strongest model for every step. For game building, I would split the workflow: 1. Fable 5 for game design + architecture 2. a fast coding model for routine implementation 3. a vision-capable model for screenshot/UI feedback 4. a cheaper model for docs, test cases, and small fixes 5. fallback when latency, cost, or output quality becomes a problem That is the real AI coding stack. Not “one magic model does everything.” More like: the right model, for the right task, at the right cost, with fallback when things break. This is why I’ve been looking at ZenMux ZenMux. ZenMux gives developers one gateway to access multiple leading AI models, with OpenAI / Anthropic / Google Vertex compatible APIs, cost tracking, quality benchmarks, auto-routing, and compensation when output quality, latency, or throughput falls short. If AI can now make games, the next question is not just “which model is strongest?” It is:how do we manage the whole model workflow Fable 5 shows the creative ceiling. ZenMux is closer to the infrastructure layer you need when AI coding becomes a real production habit.

Rachel🥥

61,143 views • 1 month ago

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning
0:12

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning

Tom Yeh

25,852 views • 17 days ago