Skip to content

Pull requests: EvolvingLMMs-Lab/lmms-eval

New pull request New
13 Open 851 Closed
13 Open 851 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

fix(distributed): call destroy_process_group() on exit to prevent NCC…
#1391 opened Jul 6, 2026 by mandanajignesh-byte Contributor Loading…
7 tasks
1
ViZDoom
#1368 opened Jun 21, 2026 by pufanyi Collaborator Loading…
7 tasks
Add Qwen-native JSON coordinate variants for pointing tasks
#1361 opened Jun 5, 2026 by njb-nvidia Contributor Loading…
1
fix: guard choices[0] and message=None before content access (41 sites, 32 files)
#1332 opened May 17, 2026 by qizwiz Loading…
1
Feat/ollama model
#1322 opened May 5, 2026 by eliasubz Loading…
1 of 7 tasks
3
feat: vLLM-Omni for video generation models
#1314 opened Apr 27, 2026 by pufanyi Collaborator • Draft
fix(evaluator): auto-init gloo process group for multi-rank launches
#1306 opened Apr 23, 2026 by Luodian Contributor • Draft
2 tasks
fix(api/task): guard empty results list in process_results
#1305 opened Apr 23, 2026 by Luodian Contributor • Draft
2 tasks
fix(eval): abort all ranks on per-rank exception instead of deadlocking
#1304 opened Apr 23, 2026 by Luodian Contributor • Draft
2 tasks
Fix missing Task import for type annotation in evaluator
#1291 opened Apr 10, 2026 by luv-oct22 Loading…
2 tasks
feat: add physics reasoning benchmarks (PhysBench, ContPhy, PhysGame, PhysicsRW, PhysReason)
#1272 opened Mar 26, 2026 by Luodian Contributor Loading…
4 tasks
feat: add VBench video generation evaluation benchmark
#1271 opened Mar 26, 2026 by Luodian Contributor Loading…
3 tasks
feat: add MiniMax as LLM judge provider (default model: MiniMax-M3)
#1263 opened Mar 22, 2026 by octo-patch Loading…
3 tasks done
2
ProTip! Follow long discussions with comments:>50.