Skip to content

Pull requests: ModelTC/LightLLM

New pull request New
48 Open 984 Closed
48 Open 984 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

qwen3 omni support long audio
#1268 opened Apr 9, 2026 by WANDY666 Contributor Loading…
2
【draft】Mtp optimization
#1266 opened Apr 9, 2026 by hiworldwzj Collaborator Loading…
6
feat: refactor kv buffer
#1265 opened Apr 8, 2026 by blueswhen Collaborator Loading…
5
fix: align /v1/chat/completions SSE stream with OpenAI spec
#1262 opened Apr 7, 2026 by sufubao Collaborator Loading…
6 tasks
2
fix error response format to match OpenAI API standard
#1259 opened Apr 7, 2026 by sufubao Collaborator Loading…
4 tasks
4
fix: run tokenizer.encode in executor to avoid blocking event loop
#1258 opened Apr 6, 2026 by sufubao Collaborator Loading…
1 of 2 tasks
3
fix: prevent orphan VIT model processes on crash or interrupt
#1256 opened Apr 3, 2026 by sufubao Collaborator Loading…
4 tasks
3
feat: add OOM pre-check for vision models and fix InternVL image dime…
#1253 opened Apr 2, 2026 by sufubao Collaborator Loading…
4
fix: improve robustness of request handling and shared memory linking
#1250 opened Apr 1, 2026 by sufubao Collaborator Loading…
3
docs: add multimodal request guide for CN and EN
#1249 opened Apr 1, 2026 by sufubao Collaborator Loading…
4
DRAFT: basic Profiler support and WIP hierarchical model level to op level profiling support
#1247 opened Apr 1, 2026 by WuSiYu Collaborator Loading…
8
feat: auto-calculate max_new_tokens to align with vLLM behavior
#1244 opened Mar 23, 2026 by sufubao Collaborator Loading…
3 tasks done
3
feat: page size > 1 support
#1224 opened Mar 10, 2026 by blueswhen Collaborator Loading…
16
fix: implement RestrictedUnpickler to mitigate CVE-2026-26220
#1219 opened Mar 5, 2026 by RinZ27 Loading…
3
feat: add multi node disk cache
#1218 opened Mar 4, 2026 by blueswhen Collaborator Loading…
4
fix: patch CVE-2026-26220
#1217 opened Mar 3, 2026 by lucrbvi Loading…
7
refactor: clean up logging and add system status reporter
#1204 opened Feb 6, 2026 by sufubao Collaborator Loading…
4
feat: add MetaX backend
#1198 opened Jan 30, 2026 by zhangts20 Loading…
5
feat: RL training support for VERL
#1196 opened Jan 28, 2026 by shihaobai Collaborator Loading…
11
feat: add pyproject.toml for uv one-click installation
#1195 opened Jan 28, 2026 by sufubao Collaborator Loading…
5
feat(R3) : Support recording expert indices for rollout router replay
#1190 opened Jan 22, 2026 by sufubao Collaborator Loading…
7
update openai_api
#1172 opened Jan 8, 2026 by WANDY666 Contributor Loading…
3
add-glm4v model
#1157 opened Dec 25, 2025 by SangChengC Contributor Loading…
9
[Feat] Add triton autotuner fallback logic
#1148 opened Dec 15, 2025 by flyinglandlord Collaborator Loading…
3
Add top_logprobs Support
#1124 opened Nov 24, 2025 by ahmadbasyouni10 Loading…
9
ProTip! What’s not been updated in a month: updated:<2026-03-11.