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
- [Bugfix] Make hc_expand/hc_contract rank-agnostic
- [ROCm][MLA] Add an AITER ASM round-robin decode route for DCP multi-token verify
- [Bugfix][Kernel] Mask block-table loads in unified attention to prevent OOB read
- [Bugfix][Kernel] Fix UnboundLocalError in triton_scaled_mm when use_heuristic=False
- [KV Offload] CPU offload capacity is divided by the number of KV cache groups on hybrid models: every `OffloadKey` (block_hash, group_idx) allocates a full multi-group chunk row
- [Bugfix] Reject empty AOT mega artifacts instead of silently running eager
- [Kimi-K3] Use the fused K/V pack kernel on the decode-context-parallel prefill path
- [Bug]: HiSparseConnector produces incoherent outputs on GLM-5.2 DEP8
- [Bug/Perf] DeepSeek-V4.1-Flash on SM120 (RTX PRO 6000 Blackwell, 8xTP): extremely low decode throughput with --enforce-eager; CUDA graphs unusable — please prioritize SM120 graph capture fix
- [KV Offload] Offload namespace records a different cache_dtype depending on the executor, so a TP1 MLA engine never shares blocks with mp engines
- Docs
- Python not yet supported