Skip to content

Pull requests: sgl-project/mini-sglang

New pull request New
20 Open 56 Closed
20 Open 56 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Feature] Enable TP size of 8 for Qwen3 MOE models
#95 opened Mar 4, 2026 by jiahe7ay Loading…
2
refactor(tests): convert to pytest-style with integration markers
#94 opened Mar 4, 2026 by MisakaVan • Draft
1
[Fix] Fix OOM during weight loading with tensor parallelism
#93 opened Mar 4, 2026 by NikitosKh Loading…
3
Add Mistral model support
#92 opened Mar 4, 2026 by NikitosKh Loading…
[Chore] Make pre-commit happy
#91 opened Mar 4, 2026 by MisakaVan Loading…
Fix: torch.AcceleratorError: CUDA error: an illegal memory access was encountered
#89 opened Mar 1, 2026 by itechbear Loading…
9
[Fix] Fix TP sampler inconsistency bug
#85 opened Feb 26, 2026 by DarkSharpness Loading…
1
2
[Feature] Support hierarchical cache
#82 opened Feb 24, 2026 by DarkSharpness Loading…
Add graph replay dump tensor tool
#72 opened Jan 30, 2026 by wlc952 Loading…
2
Adding non-streaming response (stream=False)
#69 opened Jan 20, 2026 by goswamig Loading…
2
feat: Add INT8 quantization support
#57 opened Dec 30, 2025 by louiswang524 Loading…
perf: Optimize CUDA graph batch size selection and padding
#56 opened Dec 30, 2025 by louiswang524 Loading…
feat: Implement batch tokenization for improved throughput
#55 opened Dec 30, 2025 by louiswang524 Loading…
[Refactor] Restructure test suite to match source layout and isolate benchmarks
#53 opened Dec 29, 2025 by DhiraPT Loading…
[Feature] Add MLA configuration and KV cache storage kernel
#42 opened Dec 23, 2025 by DhiraPT Loading…
1
[Education] Offline benchmark performance of Qwen3-0.6B on MLX (CPU) and Modal (GPU)
#40 opened Dec 23, 2025 by lamng3 Loading…
[Feature] Implement variable page size support
#33 opened Dec 22, 2025 by DhiraPT Loading…
21
docs: align README/features CLI examples with args.py
#29 opened Dec 21, 2025 by Taskrwu Loading…
1
[Improvement] Enhance engine error handling and documentation add more logging and doc
#23 opened Dec 20, 2025 by louiswang524 Loading…
Request-scoped Torch profiler via profile flag.
#14 opened Dec 18, 2025 by AdamLouly Loading…
@DarkSharpness
1
ProTip! Add no:assignee to see everything that’s not assigned.