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
- [CI] [AMD] pr-test: Run stage-b and diffusion shards in parallel; trim stage-a
- [Perf] mem_cache: Avoid O(R) list shifts and free_pages scans in RangePageAllocator
- [Perf] dsv41: Fuse decode mHC post, combine and RMSNorm into one kernel
- [Perf] dsv41: Add a row floor for the Mega-mHC prefill boundary
- [Perf] moe: Fold the routed scaling factor into the mega MoE pre-dispatch
- [Perf] dsv41: Add replicated and hybrid Engram host-table layouts
- [Perf] moe: Pass the global token count to the Mega MoE tile heuristic
- [Feature] Semantic Prompt Hints in HiCache TreeNode for Heterogeneous L3 Storage Tiering
- [Perf] diffusion: Hash large conditioning-cache keys in parallel chunks
- [Feat] scheduler: add optional consecutive-prefill resource budgets
- Docs
- Python not yet supported