Skip to content

Actions: modelscope/mcore-bridge

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
742 workflow runs
742 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
fix: support QSA with sequence parallelism (#180) Lint test #732: Commit c419b4b pushed by tastelikefeet
34s main
main
34s
fix num_samples attributes error(#186) Lint test #731: Commit e16516e pushed by tastelikefeet
27s main
main
27s
fix num_samples attributes error Lint test #730: Pull request #186 opened by tastelikefeet
27s fix/num_sample
fix/num_sample
27s
fix Lint test #729: Commit f36d108 pushed by tastelikefeet
34s fix/num_sample
fix/num_sample
34s
optimize qwen3.8-flash-next Lint test #728: Pull request #185 synchronize by hjh0119
32s hjh0119:qwen4-0903
hjh0119:qwen4-0903
32s
optimize qwen3.8-flash-next Lint test #727: Pull request #185 opened by hjh0119
37s hjh0119:qwen4-0903
hjh0119:qwen4-0903
37s
fix(qwen4_exp): fall back QSA to full attention under sequence parallelism Lint test #726: Pull request #184 opened by ehooon
1d 5h 6m 31s ehooon:fix-qwen4-exp-qsa-sp
ehooon:fix-qwen4-exp-qsa-sp
1d 5h 6m 31s
Fix two FP8-training issues: PLE ngram dequantization and LoRA fp8 autocast guard Lint test #725: Pull request #183 opened by ehooon
Action required ehooon:fix-fp8-training
ehooon:fix-fp8-training
Action required
Fix packed sample count derivation Lint test #724: Pull request #181 opened by taking-lying-flat
3d 19h 8m 59s taking-lying-flat:fix/packed-seq-num-samples
taking-lying-flat:fix/packed-seq-num-samples
3d 19h 8m 59s
fix: support QSA with sequence parallelism Lint test #723: Pull request #180 synchronize by taking-lying-flat
36s taking-lying-flat:fix/qsa-sequence-parallel
taking-lying-flat:fix/qsa-sequence-parallel
36s
fix: support QSA with sequence parallelism Lint test #722: Pull request #180 synchronize by taking-lying-flat
4d 4h 38m 25s taking-lying-flat:fix/qsa-sequence-parallel
taking-lying-flat:fix/qsa-sequence-parallel
4d 4h 38m 25s
fix: support QSA with sequence parallelism Lint test #721: Pull request #180 opened by taking-lying-flat
4d 4h 51m 42s taking-lying-flat:fix/qsa-sequence-parallel
taking-lying-flat:fix/qsa-sequence-parallel
4d 4h 51m 42s
[bugfix] preserve QSA MRoPE batch dimension Lint test #720: Pull request #179 opened by taking-lying-flat
30s taking-lying-flat:fix/qsa-mrope-batch-dimension
taking-lying-flat:fix/qsa-mrope-batch-dimension
30s
feat(model): add generic chunked linear cross entropy Lint test #719: Pull request #177 opened by maomaocun
Action required maomaocun:codex/chunked-linear-ce-h100
maomaocun:codex/chunked-linear-ce-h100
Action required
[model] refactor qwen3.5 GDN to use FLA causal conv1d and cu_seq_lens_q Lint test #718: Pull request #176 synchronize by ys2025-AI
31s ys2025-AI:main
ys2025-AI:main
31s
[model] refactor qwen3.5 GDN to use FLA causal conv1d and cu_seq_lens_q Lint test #717: Pull request #176 synchronize by ys2025-AI
29s ys2025-AI:main
ys2025-AI:main
29s
[model] refactor qwen3.5 GDN to use FLA causal conv1d and cu_seq_lens_q Lint test #716: Pull request #176 opened by ys2025-AI
33s ys2025-AI:main
ys2025-AI:main
33s
Fix packed sequence length handling for Megatron-Core 0.16.0/0.16.1 (… Lint test #715: Commit 4f2a95c pushed by tastelikefeet
32s main
main
32s
Fix packed sequence length handling for Megatron-Core 0.16.0/0.16.1 Lint test #714: Pull request #175 synchronize by hazelduan
34s hazelduan:gdn_model_hybrid
hazelduan:gdn_model_hybrid
34s
Fix packed sequence length handling for Megatron-Core 0.16.0/0.16.1 Lint test #713: Pull request #175 opened by hazelduan
30s hazelduan:gdn_model_hybrid
hazelduan:gdn_model_hybrid
30s
wip Lint test #712: Commit 27ea13c pushed by tastelikefeet
41s swift-dev-v5
swift-dev-v5
41s
[model] qwen3.8-flash-next (#174) Lint test #711: Commit 09e891a pushed by tastelikefeet
30s main
main
30s
[model] qwen3.8-flash-next Lint test #710: Pull request #174 synchronize by hjh0119
38s hjh0119:qwen4
hjh0119:qwen4
38s
[model] qwen3.8-flash-next Lint test #709: Pull request #174 synchronize by hjh0119
31s hjh0119:qwen4
hjh0119:qwen4
31s
[model] qwen3.8-flash-next Lint test #708: Pull request #174 opened by hjh0119
27s hjh0119:qwen4
hjh0119:qwen4
27s