PrototypeCommon
harleyszhang/rapid_llm
RapidLLM is an LLM inference framework with continuous batching, tensor and data parallelism, quantized weights, and with pluggable Triton/CUDA kernels.
- Owner
- harleyszhang (personal account)
- Stars
- Stars: 196
- Forks
- Forks: 33
- Language
- Python
- License
- None
- Created
- 2024
- Topics
- llama3llmllm-inferencepython3triton-kernelsattentioncontinuescontinuous-batchingflash-attention-3qwen3-moe
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by harleyszhang
All 5 repos by harleyszhang- harleyszhang/cv_noteStars: 2,651PythonRC
- harleyszhang/llm_noteStars: 896PythonBeta
- harleyszhang/dl_noteStars: 532PythonAlpha
- harleyszhang/llm_countsStars: 125PythonPrototype
Nearby in Python
All Python repos- dacnay816y62-hub/fantasy-movie-poster-skillStars: 196PythonPrototype
- zhezh/kddcup2026_championStars: 196PythonPrototype
- Jakeschincariol/chatgpt-youtube-agent-skillStars: 196PythonPrototype
- patpadgett/corkscrewStars: 196PythonPrototype
- wbond/sublime_prefixrStars: 196PythonPrototype
- seomoz/reppyStars: 196PythonPrototype