Skip to content

Actions: hpcaitech/ColossalAI

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
2,117 workflow runs
2,117 workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
[hotfix] fix parameter shape checking issue-translator #6112: Issue comment #6124 (comment) created by JazyJiang
Queued
Queued
Fix: Handle FusedLAMB Fails with LowLevelZeroPlugin When Using CPU Offload issue-translator #6111: Issue comment #6418 (comment) created by JazyJiang
Queued
Queued
fix: add path validation in profiling_utils.py... issue-translator #6110: Issue comment #6430 (comment) created by JazyJiang
Queued
Queued
[DOC]: Update the documentation of ShardConfig for 1D, 2D, 2.5D, 3D tensor parallelism issue-translator #6109: Issue comment #6278 (comment) created by JazyJiang
Queued
Queued
Fix: Handle LowLevelZeroPlugin with use_fp8=True issue-translator #6108: Issue comment #6419 (comment) created by JazyJiang
Queued
Queued
Close inactive issues Close inactive issues #1709: Scheduled
1s main
main
1s
Synchronize Submodule Synchronize Submodule #1694: Scheduled
Queued main
main
Queued
Build Documentation On Schedule & After Release Build Documentation On Schedule & After Release #1241: Scheduled
Queued main
main
Queued
ci: fix layered GPU batch compatibility CI / GPU Idle Monitor #4: Commit 08c413e pushed by richardoo-707
2h 35m 13s ci/three-layer-workflow-validation
ci/three-layer-workflow-validation
2h 35m 13s
ci: trigger self-contained GPU probe validation CI / GPU Idle Monitor #3: Commit 6f2ca60 pushed by richardoo-707
1h 14m 50s ci/three-layer-workflow-validation
ci/three-layer-workflow-validation
1h 14m 50s
Merge pull request #6442 from hpcaitech/ci/e1-runner-bootstrap .github/workflows/example_check_on_dispatch.yml #149: Commit be1a04a pushed by yuxuan-lou
Failure main
main
Failure
Close inactive issues Close inactive issues #1708: Scheduled
1s main
main
1s
Synchronize Submodule Synchronize Submodule #1693: Scheduled
1d 0h 0m 11s main
main
1d 0h 0m 11s
[shardformer]: fix llama causal mask with sequence parallelism and no flash attention Build on PR #9380: Pull request #6452 opened by LiRunGuo
Action required LiRunGuo:fix/llama-sp-causal-mask
LiRunGuo:fix/llama-sp-causal-mask
Action required
[shardformer]: fix attention mask of qwen2/qwen3 with pipeline + sequence parallelism Build on PR #9379: Pull request #6451 opened by LiRunGuo
Action required LiRunGuo:fix/qwen-pp-sp-attn-mask
LiRunGuo:fix/qwen-pp-sp-attn-mask
Action required
[shardformer]: support Qwen3-MoE with expert parallelism Build on PR #9378: Pull request #6450 opened by LiRunGuo
Action required LiRunGuo:feat/qwen3-moe
LiRunGuo:feat/qwen3-moe
Action required
[checkpoint]: unpad padded parameters when saving sharded MoE checkpoints Build on PR #9377: Pull request #6449 opened by LiRunGuo
Action required LiRunGuo:fix/moe-ckpt-unpad
LiRunGuo:fix/moe-ckpt-unpad
Action required
[tensor]: fix saving checkpoints of padded parameters with tp_size=1 (keep detach/clone of padded tensors unpaddable) Build on PR #9376: Pull request #6448 opened by LiRunGuo
Action required LiRunGuo:fix/padded-tensor-detach
LiRunGuo:fix/padded-tensor-detach
Action required
Build Documentation On Schedule & After Release Build Documentation On Schedule & After Release #1240: Scheduled
1d 0h 0m 2s main
main
1d 0h 0m 2s
ci: make hourly GPU probe lightweight CI / GPU Idle Monitor #2: Commit 4d0f379 pushed by richardoo-707
2m 24s ci/three-layer-workflow-validation
ci/three-layer-workflow-validation
2m 24s
ci: retry GPU layers when capacity is idle CI / GPU Idle Monitor #1: Commit aa98c62 pushed by richardoo-707
7m 6s ci/three-layer-workflow-validation
ci/three-layer-workflow-validation
7m 6s
ci: trigger layer 3 validation CI / Layer 3 / Full GPU Matrix #1: Commit 2c1d55a pushed by richardoo-707
6m 18s ci/three-layer-workflow-validation
ci/three-layer-workflow-validation
6m 18s
Close inactive issues Close inactive issues #1707: Scheduled
1s main
main
1s
Synchronize Submodule Synchronize Submodule #1692: Scheduled
1d 0h 0m 3s main
main
1d 0h 0m 3s
[gemini]: fix loading checkpoints of models with tied parameters Build on PR #9375: Pull request #6447 opened by LiRunGuo
Action required LiRunGuo:fix/gemini-load-tied-weights
LiRunGuo:fix/gemini-load-tied-weights
Action required