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 supported35 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Gemma4: don't crash when a KV-shared layer has no same-type target
- [Frontend] Add offline prefix-cache workload analyzer CLI
- [Bug]: prompt_logprobs argmax disagrees with greedy decode for sequences ≥256 tokens — num_tokens-dependent block size in RMSNorm kernels defeats VLLM_BATCH_INVARIANT
- [Bug][Anthropic] stop_sequence never populated + unguarded json.loads crash in messages_full_converter
- [Bug][Responses API] Zero-delta streaming items skip added/in_progress lifecycle events
- Platform plugin registration failures are silently swallowed, causing undiagnosed CPU fallback
- [Feature]: Folding per-platform default arguments into the vLLM engine
- DeepseekV32IndexerBackend requires --block-size 64 — not documented or auto-detected
- [Bug]: llama3_json, jamba and ernie45 tool parsers drop tool calls when the whole message arrives in a single streaming delta
- [RFC] LoRA Adapter Lifecycle for RL Training
- Docs
- Python not yet supported