Skip to content

Actions: hao-aaron/SkyRL

Actions

SkyRL-GPU

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
gpu_skyrl.yaml
18 workflow runs
18 workflow runs

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][train] Fix port collision for num_engines > 1 in non-PD case (#… SkyRL-GPU #18: Commit c2d6d47 pushed by hao-aaron
27s main
main
27s
[fix][CI] Fix cleanup for test_weight_sync (#1510) SkyRL-GPU #17: Commit 1f8c6f2 pushed by hao-aaron
22s main
main
22s
[bug] Fix logprobs handling and error status codes for vllm-router co… SkyRL-GPU #16: Commit 791fbf9 pushed by hao-aaron
23s main
main
23s
[megatron][checkpointing] fix checkpointing with optimizer cpu offloa… SkyRL-GPU #15: Commit e62cf3e pushed by hao-aaron
26s main
main
26s
[bug] fix sleep bugs in new inference codepath (#1383) SkyRL-GPU #14: Commit 81e342c pushed by hao-aaron
25s main
main
25s
[dep] Bump vLLM to 0.18 (#1374) SkyRL-GPU #13: Commit d00b17e pushed by hao-aaron
30s main
main
30s
x SkyRL-GPU #12: Commit 3e808b1 pushed by hao-aaron
33s main
main
33s
x SkyRL-GPU #11: Commit 75ba8cd pushed by hao-aaron
25s main
main
25s
[tx] Optimize the decode performance (#1363) SkyRL-GPU #10: Commit 58e3ab7 pushed by hao-aaron
22s main
main
22s
[bug][algorithm] remove incorrect torch.no_grad() for kl in loss (use… SkyRL-GPU #9: Commit 9d7bca9 pushed by hao-aaron
22s main
main
22s
[train] Add worker_process_setup_hook to set mp start method to `sp… SkyRL-GPU #8: Commit dffae95 pushed by hao-aaron
25s main
main
25s
[fix] Fix placement group bundle ordering for inference engines (#1308) SkyRL-GPU #7: Commit 797b57c pushed by hao-aaron
23s main
main
23s
[train][1/N] Native Weight Sync API: NCCL (#1271) SkyRL-GPU #6: Commit dfdee82 pushed by hao-aaron
22s main
main
22s
[trivial] Fix comments numbering and make some code more concise in t… SkyRL-GPU #5: Commit 2fb789a pushed by hao-aaron
27s main
main
27s
[ci] remove incorrect key check in megatron ci test (#1275) SkyRL-GPU #4: Commit 2f40a07 pushed by hao-aaron
32s main
main
32s
[tx] Implement Qwen 3.5 model architecture (#1228) SkyRL-GPU #3: Commit adcb93e pushed by hao-aaron
24s main
main
24s
Add GLM-4.7-Flash (30B MoE) training example (#1215) SkyRL-GPU #2: Commit 21660da pushed by hao-aaron
32s main
main
32s
[deps] remove numpy from override dependencies (#1178) SkyRL-GPU #1: Commit d1e263b pushed by hao-aaron
23s main
main
23s