All Tools
J
LLMFreeOpen Source
JITTORLLMS
Low-resource LLM inference library on Jittor
Apache-2.0
ABOUT
Running open LLMs often needs high VRAM and CUDA-only runtimes. JittorLLMs serves models on the Jittor JIT stack with lower hardware requirements and stronger Chinese support so teams infer locally without a full PyTorch serving stack.
INTEGRATION GUIDE
1. Run open Chinese LLMs locally with a Jittor inference API
2. Serve language models on machines that cannot host vLLM
3. Port Jittor research models into a shared LLM runtime
4. Benchmark low-resource inference against CUDA-only servers
TAGS
pythonllminferencejittorchineseserving