Skip to content

Pull requests: sgl-project/mini-sglang

New pull request New
29 Open 81 Closed
29 Open 81 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] Vanilla speculative decoding (standalone, opt-in offline interface)
#134 opened May 24, 2026 by javierlimt6 Loading…
5 tasks done
4
[Feature] Support Qwen3.5 text model
#133 opened May 20, 2026 by botieking98 Loading…
[Feature] Add optional FP8 (float8_e4m3fn) KV cache pool
#132 opened May 16, 2026 by javierlimt6 Loading…
6 of 7 tasks
7
[Feature] Add usage token counting and finish_reason to API responses observability Tools and changes for debugging, profiling, tracing, logging, metrics, and runtime diagnostics
#126 opened May 8, 2026 by abinggo Contributor Loading…
2
Feature/support more sampling parameters enhancement New feature or request
#121 opened Apr 20, 2026 by Alise-svg Loading…
1
feat: add configurable scheduling policy enhancement New feature or request
#119 opened Apr 20, 2026 by Alise-svg Loading…
1
WIP: add text gemma3 support enhancement New feature or request
#118 opened Apr 19, 2026 by YzXiao101 Contributor • Draft
feat: implement batch tokenization for TokenizeManager duplicate This issue or pull request already exists enhancement New feature or request
#117 opened Apr 19, 2026 by Alise-svg Loading…
3
Feat: Enable structured output in Mini-SGLang enhancement New feature or request
#115 opened Apr 2, 2026 by YzXiao101 Contributor Loading…
4 of 6 tasks
Add text-only Llama 4 support to mini-sglang. enhancement New feature or request
#114 opened Apr 1, 2026 by sheepfish5 Loading…
[Fix] Avoid double-free in overlap_loop if Req is aborted while finishing bugfix Fixes incorrect behavior, runtime errors, or regressions.
#111 opened Mar 25, 2026 by MisakaVan Contributor Loading…
1
[refactor] enable page size > 1 for fi backend enhancement New feature or request
#110 opened Mar 24, 2026 by zzh-stable Loading…
1
Add request-scoped profiler benchmark flag and compressed trace export observability Tools and changes for debugging, profiling, tracing, logging, metrics, and runtime diagnostics
#109 opened Mar 21, 2026 by CrazyDave999 Loading…
Optimize load_weight with per-file batch H2D and zero-copy CPU sharding misc Minor maintenance changes that do not alter core behavior.
#108 opened Mar 20, 2026 by staryxchen Contributor Loading…
[Feture] Add reasoning-parser enhancement New feature or request
#107 opened Mar 19, 2026 by jiahe7ay Contributor Loading…
1
[Fix]Pad the last rank if vocab size is not divisible by tp_size bugfix Fixes incorrect behavior, runtime errors, or regressions.
#100 opened Mar 9, 2026 by cswuyg Contributor Loading…
1
[Feature] Better estimation policy enhancement New feature or request
#97 opened Mar 8, 2026 by YzXiao101 Contributor Loading…
7 tasks done
5
[Feature] Expert parallelism support for MoE models enhancement New feature or request
#96 opened Mar 6, 2026 by NikitosKh Contributor Loading…
1
refactor(tests): convert to pytest-style with integration markers tests Tests, benchmarks, and other checks for validating correctness, performance, or regressions.
#94 opened Mar 4, 2026 by MisakaVan Contributor • Draft
1
[Feature] Support hierarchical cache enhancement New feature or request poc Proof-of-concept or experimental changes that are not intended for merge.
#82 opened Feb 24, 2026 by DarkSharpness Collaborator Loading…
Add graph replay dump tensor tool observability Tools and changes for debugging, profiling, tracing, logging, metrics, and runtime diagnostics
#72 opened Jan 30, 2026 by wlc952 Loading…
2
feat: Add INT8 quantization support enhancement New feature or request
#57 opened Dec 30, 2025 by louiswang524 Contributor Loading…
perf: Optimize CUDA graph batch size selection and padding misc Minor maintenance changes that do not alter core behavior.
#56 opened Dec 30, 2025 by louiswang524 Contributor Loading…
feat: Implement batch tokenization for improved throughput enhancement New feature or request
#55 opened Dec 30, 2025 by louiswang524 Contributor Loading…
[Refactor] Restructure test suite to match source layout and isolate benchmarks tests Tests, benchmarks, and other checks for validating correctness, performance, or regressions.
#53 opened Dec 29, 2025 by DhiraPT Loading…
ProTip! Mix and match filters to narrow down what you’re looking for.