vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: Recurring Xid 13 chip-wide warp errors (misaligned address / illegal instruction / out-of-range register) on SM120 under sustained multi-hour load — Nemotron-3.5-Lightning-30B-A3B-NVFP4, marlin MoE, hybrid Mamba
- [Bugfix][V1] Restore hybrid GDN prefix-cache hits under MTP spec decoding
- [Bug]: DeepSeek-V4 encoder emits an empty `<think></think>` block for assistant turns without reasoning
- Decode optimizations for NVIDIA confidential computing
- [Model][Generative Recommendation] Add HSTU model support to vLLM
- [Bug]: All GPTQ checkpoints fail with UR_RESULT_ERROR_DEVICE_LOST on Intel Arc B60 (XPU) at profile_run
- [Feature] Support batch invariance on ROCm
- fix(docs): correct "Mixture-of-Expert" to "Mixture-of-Experts" in README
- feat(inkling): add Eagle3 hidden state extraction support
- [Bug]: ImportError: libcudart.so.13: cannot open shared object file: No such file or directory
- Docs
- Python not yet supported