Skip to content

Pull requests: vllm-project/vime

New pull request New
23 Open 255 Closed
23 Open 255 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

[Bugfix][Training] Avoid entropy grad graph when entropy_coef is zero
#330 opened Jul 9, 2026 by Henry-Jessie Loading…
1
[bugfix] Fix distributed update weights for pipeline parallel
#329 opened Jul 8, 2026 by knlnguyen1802 Collaborator Loading…
6
Add on-policy distillation example (Qwen3-4B + Qwen3-32B vLLM teacher)
#328 opened Jul 8, 2026 by CalvinXKY Collaborator • Draft
3 tasks
6
fix(docker): patch vLLM v0.23.0 to preserve non-persistent buffers on layerwise reload
#326 opened Jul 8, 2026 by Josephasafg Contributor Loading…
2
[NPU] replace 30b-npu script with ray-based launch method
#323 opened Jul 8, 2026 by floatlibai Loading…
8
feat(rollout): add TransferQueue trajectory ingestion for closed-loop VIME training
#320 opened Jul 7, 2026 by aoshen02 Collaborator • Draft
8
fix(npu): adapt colocate patch for vllm-ascend and vllm version upgrade
#318 opened Jul 3, 2026 by miracle0517 Loading…
2 tasks done
5
【NPU】Added patch update verification for NPU CI bug Something isn't working
#316 opened Jul 2, 2026 by yzeyu71 Loading…
8
[NPU] Enable Qwen3-VL GRPO (single- & multi-turn) on Ascend
#314 opened Jul 2, 2026 by Meihan-chen Collaborator • Draft
2
[Bugfix][Rollout] Validate batched RM reward lengths
#313 opened Jul 2, 2026 by BreezyB1n Loading…
1
4
[NPU] Add Search-R1 example
#311 opened Jul 2, 2026 by Fulin-Gao Loading…
22
[WIP]fix(npu): trim vllm-ascend colocate patch and pin native IPC engine
#310 opened Jul 2, 2026 by CalvinXKY Collaborator • Draft
2 of 3 tasks
3
[Bugfix] Make reward CPU tests collectable without optional runtime deps
#309 opened Jul 1, 2026 by FeathBow Loading…
1
docker: rebase vllm.patch onto vLLM 0.24.0 (drop sleep hunk, add abort+usage)
#308 opened Jul 1, 2026 by aoshen02 Collaborator Loading…
2
[Rollout] feat: add Mooncake Store rollout transfer for disaggregated training.
#300 opened Jun 30, 2026 by stmatengss Loading…
3 tasks done
13
[NPU] support multi-agent rollout pipeline
#298 opened Jun 30, 2026 by Windfeng8 Loading…
12
[NPU]: support use-rollout-routing-replay
#292 opened Jun 23, 2026 by Windfeng8 Loading…
6
Npu rebase
#261 opened Jun 17, 2026 by momo609 Collaborator Loading…
14
feat(coding_agent_rl): add SWE-bench harness evaluation + uniagent mode
#250 opened Jun 15, 2026 by aoshen02 Collaborator Loading…
3 tasks
3
[Feature]: Lora support
#228 opened Jun 10, 2026 by princepride Collaborator Loading…
6
[NPU] support qwen3-30b model.
#189 opened Jun 8, 2026 by momo609 Collaborator Loading…
10
[Feature] shard-level P2P weight transfer (Phase 1, #160)
#161 opened Jun 5, 2026 by CalvinXKY Collaborator Loading…
6 tasks
6
[1/N] Add optional NCCL m2n weight sync backend
#159 opened Jun 5, 2026 by princepride Collaborator Loading…
6
ProTip! Type g i on any issue or pull request to go back to the issue listing page.