#cuda-graphs Repositories
Discover 5 active open source GitHub repositories tagged with #cuda-graphs, accumulating 4,010 total stars (averaging 802 stars per project). Leading this topic is Lightning-AI/lightning-thunder with 1,476 stars.
5 repositories total
PyTorch compiler that accelerates training and inference. Get built-in optimizations for performance, memory, parallelism, and easily write your own.
LeanRL is a fork of CleanRL, where selected PyTorch scripts optimized for performance using compile and cudagraphs.
WeDLM: The fastest diffusion language model with standard causal attention and native KV cache compatibility, delivering real speedups over vLLM-optimized baselines.
Fast, exact inference for StartLux-Decision models on NVIDIA GPUs: the released GGUF files run on int8 and 4-bit tensor cores behind the /v1/systemone API, delivering up to 5.3ร lower latency than vLLM in FP8 with bit-identical answers.
Frequently Asked Questions
What is the most popular #cuda-graphs repository on GitHub?
The top ranked open source #cuda-graphs project on GitHub is Lightning-AI/lightning-thunder with 1,476 stars. PyTorch compiler that accelerates training and inference. Get built-in optimizations for performance, memory, parallelism, and easily write your own.
Which programming languages are most used in #cuda-graphs repositories?
Among the top ranked #cuda-graphs projects on GitHub, the most common programming languages are Python, C++.
How is the #cuda-graphs repository ranking calculated?
We track 5 active open source repositories tagged with #cuda-graphs (500+ stars), totaling 4,010 stars with an average of 802 stars per repository. Rankings update daily from the GitHub API.