Skip to content

Actions: THUDM/slime

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
6,054 workflow runs
6,054 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[Bug] UpdateWeightFromDistributed fails for VLM models in non-colocate mode Slash Command Handler #423: Issue comment #1673 (comment) created by ZHZisZZ
10s
10s
Featuring: Convert tool for Qwen3-VL moe model to hf format after training with Megatron Slash Command Handler #422: Issue comment #1510 (comment) created by coding-famer
7s
7s
[Bug] UpdateWeightFromDistributed fails for VLM models in non-colocate mode Slash Command Handler #421: Issue comment #1673 (comment) created by coding-famer
7s
7s
[Feature] Add modular tracking interface with MLflow backend Slash Command Handler #420: Issue comment #1591 (comment) created by Lawhy
6s
6s
[Bug] Actor unavailable error Slash Command Handler #419: Issue comment #1739 (comment) created by zengxingchen
7s
7s
[Bug] UpdateWeightFromDistributed fails for VLM models in non-colocate mode Slash Command Handler #418: Issue comment #1673 (comment) created by ZHZisZZ
8s
8s
Featuring: Convert tool for Qwen3-VL moe model to hf format after training with Megatron Slash Command Handler #417: Issue comment #1510 (comment) edited by cjy0x
9s
9s
[Bug] UpdateWeightFromDistributed fails for VLM models in non-colocate mode Slash Command Handler #416: Issue comment #1673 (comment) created by cjy0x
7s
7s
Support qwen3.5 loss mask for multi-turn SFT conda CI #2849: Pull request #1742 synchronize by huang3eng
Action required huang3eng:main
huang3eng:main
Action required
Support qwen3.5 loss mask for multi-turn SFT PR Test #1498: Pull request #1742 synchronize by huang3eng
Action required huang3eng:main
huang3eng:main
Action required
Support qwen3.5 loss mask for multi-turn SFT pre-commit #3337: Pull request #1742 synchronize by huang3eng
Action required huang3eng:main
huang3eng:main
Action required
[docker] update sglang patch (#1743) pre-commit #3336: Commit a1ee76c pushed by zhuzilin
50s main
main
50s
[docker] update sglang patch pre-commit #3335: Pull request #1743 opened by zhuzilin
50s docker
docker
50s
[docker] update sglang patch conda CI #2848: Pull request #1743 opened by zhuzilin
1s docker
docker
1s
Support qwen3.5 loss mask for multi-turn SFT Slash Command Handler #415: Issue comment #1742 (comment) created by huang3eng
11s
11s
Support qwen3.5 loss mask for multi-turn SFT pre-commit #3334: Pull request #1742 opened by huang3eng
Action required huang3eng:main
huang3eng:main
Action required
Support qwen3.5 loss mask for multi-turn SFT conda CI #2847: Pull request #1742 opened by huang3eng
Action required huang3eng:main
huang3eng:main
Action required
[Bug] qwen35 running with pp error Slash Command Handler #414: Issue comment #1713 (comment) edited by sys-reasoner
9s
9s
[Bug] qwen35 running with pp error Slash Command Handler #413: Issue comment #1713 (comment) created by sys-reasoner
10s
10s
size not match error after many training steps Slash Command Handler #412: Issue comment #1076 (comment) created by Superskyyy
9s
9s
(fix):not have encoder_only attr cause run failed pre-commit #3333: Pull request #1741 opened by wangyufak
Action required wangyufak:fix_sglang_engine_file
wangyufak:fix_sglang_engine_file
Action required
(fix):not have encoder_only attr cause run failed conda CI #2846: Pull request #1741 opened by wangyufak
Action required wangyufak:fix_sglang_engine_file
wangyufak:fix_sglang_engine_file
Action required
[docker] fix qwen3_vl visual module loading Slash Command Handler #411: Issue comment #1727 (comment) created by ngothetai
7s
7s
OOM when training with long sequences despite using dynamic batch size and sequence parallel Slash Command Handler #410: Issue comment #1522 (comment) created by jymmmmm
7s
7s
OOM when training with long sequences despite using dynamic batch size and sequence parallel Slash Command Handler #409: Issue comment #1522 (comment) created by Ariya12138
8s
8s