sglang
https://github.com/sgl-project/sglang
Python
SGLang is a high-performance serving framework for large language models and multimodal models.
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 supported1 Subscribers
Add a CodeTriage badge to sglang
Help out
- Issues
- [Bug] PD prefill: early cached-prefix send can transfer KV before the HiCache load-back finishes (stale KV on decode)
- [Fix] Align conv layout when copying between mamba checkpoint pool an…
- [Bug] bench_serving: --request-rate falls further behind schedule with every request, so the achieved rate is below the requested rate
- fix(openai): include cache details in continuous Completions usage
- [Bug] Transcription multipart upload has no application-level size limit
- [Fix] multimodal_gen: Fix dynamic-batching bugs - non-output-rank split, OOM memory climb
- [RFC] Monokernel framework for SGLang
- fix: normalize host logprob rows in scheduler
- [Fix] Pass return_dict=False to apply_chat_template(tokenize=True) for transformers 5
- [Feat] comfyui: Add SGLDWorkerStatus diagnostics node
- Docs
- Python not yet supported