jmaczan/torch-webgpu
PyTorch compiler and WebGPU runtime
This project helps machine learning engineers and researchers run PyTorch models, especially large language models (LLMs), on hardware that supports WebGPU. You provide your existing PyTorch model code, and it compiles and runs it efficiently using WebGPU, which is broadly supported across various devices and browsers. The result is your model executing and producing outputs without needing specialized GPU backends.
Available on PyPI.
Use this if you are a machine learning engineer or researcher who needs to deploy PyTorch models, particularly large language models, across a wide range of hardware using WebGPU as a unified runtime.
Not ideal if your primary concern is absolute peak performance on highly specialized hardware like NVIDIA CUDA GPUs, as current performance is not yet fully optimized.
Stars
14
Forks
1
Language
C++
License
—
Category
Last pushed
Feb 06, 2026
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/jmaczan/torch-webgpu"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
thu-pacman/chitu
High-performance inference framework for large language models, focusing on efficiency,...
sophgo/LLM-TPU
Run generative AI models in sophgo BM1684X/BM1688
NotPunchnox/rkllama
Ollama alternative for Rockchip NPU: An efficient solution for running AI and Deep learning...
Deep-Spark/DeepSparkHub
DeepSparkHub selects hundreds of application algorithms and models, covering various fields of...
howard-hou/VisualRWKV
VisualRWKV is the visual-enhanced version of the RWKV language model, enabling RWKV to handle...