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
- [Agentic Inference] Programmatic KV Cache for Agentic Workloads
- [NPU] Support mlaprolog, fp8 KVcache(only MLA) and ChunkPrefill for A5(950PR/DT) NPU
- [LoRA] Support LoRA under DP attention: idle-forward guards, attn-TP-local slicing and buffer sharding
- [Speculative] Seed rejection-sampling draft proposals for deterministic inference
- [HiCache] Add KVCR as an L3 storage backend for hint-driven cross-instance KV reuse
- [weight_cache] Key daemon paths by config digest
- Separate the model instance create from the load weights to improve robust
- Modified the moe_align_block method in sgl-kernel to uniform handling of invalid expert_id
- [Qwen3-Next] Add cutedsl decode/mtp kernel with transposed ssm_state and prefill gluon kernel for blackwell.
- [Fix][Perf] Improve Elastic EP fault tolerance with direct controller routing and async active-rank snapshots
- Docs
- Python not yet supported