To
TokenHawk, WebGPU Running LLaMA
TokenHawk, WebGPU Running LLaMA
Hi HN. TokenHawk is a WebGPU app that runs LLaMA models. It works on the CLI through Dawn, or directly on the web.
Share cardActual performance
2points
Did not reach leaderboard
Launch Intel predictions
Analyze your own launch →91%91% predicted probability of success on Product Hunt, based on ML models trained on real launch data.
best fitHighest predicted score across all platforms for this description.
63%63% predicted probability of success on Hacker News, based on ML models trained on real launch data.
nativeThis product was originally launched on this platform.
57%57% predicted probability of success on Indie Hackers, based on ML models trained on real launch data.
52%52% predicted probability of success on TrustMRR, based on ML models trained on real launch data.
48%48% predicted probability of success on AppSumo, based on ML models trained on real launch data.
34%34% predicted probability of success on BetaList, based on ML models trained on real launch data.
24%24% predicted probability of success on Acquire.com, based on ML models trained on real launch data.
Incorrect prediction on native model
Similar products
Ll
Llama Running on a Microcontroller76%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Llama Running on a Microcontroller
Ll
Llama or Alpaca?78%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Llama or Alpaca?
Ll
Llama 3.2 Interpretability with Sparse Autoencoders74%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Llama 3.2 Interpretability with Sparse Autoencoders
Ll
Llama 2 Uncensored 70B as API79%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Llama 2 Uncensored 70B as API
Fi
Finetune Llama-3.1 2x faster in a Colab74%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Finetune Llama-3.1 2x faster in a Colab
Ru
Running WebAssembly on ARM67%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Running WebAssembly on ARM
Ob
Oberon System Running on LuaJIT52%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Oberon System Running on LuaJIT
Ha
Hash functions from C++ running in WebAssembly67%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Hash functions from C++ running in WebAssembly
Ru
Running for stuck to chair programmers44%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Running for stuck to chair programmers
Lu
Luakernel – Lua + SQLite + musl libc running on x8689%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.
Luakernel – Lua + SQLite + musl libc running on x86