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
- [Fix] session: Prefill a retracted one-token first turn as a new request
- [Fix] diffusion: Handle seed lists in multi-output dynamic batches
- [Fix] conversation: Merge consecutive same-role turns in LLAMA2-style templates
- [Fix] conversation: Open the first user turn with [INST] after a Mistral system prompt
- [Feat] [NPU] kimi-k3: Enable prefill context parallelism (full-sequence CP) on Ascend
- [Fix] comfyui: Reject conditioning kwargs the SGLD adapters silently drop
- [Fix] RL: Fit output store reads to Mooncake's read buffer and lease
- [Feat] rust-server: multimodal (image) preprocessing through sglang-processor
- [Fix] spark2_5: Honor gate_attn_act_mode instead of hardcoding sigmoid
- [Bug] minimax-m2 streaming: next invoke's parameters overwrite the current call's arguments when one delta spans </invoke>
- Docs
- Python not yet supported