Skip to content

Actions: verl-project/verl

Actions

e2e_ppo_trainer_megatron_sglang

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
e2e_ppo_trainer_megatron_sglang.yml
5,185 workflow runs
5,185 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[rollout] feat: support auto resume on abort in FullyAsyncLLMServerManager e2e_ppo_trainer_megatron_sglang #7061: Pull request #5430 synchronize by wuxibin89
Queued wuxibin89:wuxibin/resume_client
wuxibin89:wuxibin/resume_client
Queued
[rollout] feat: support auto resume on abort in FullyAsyncLLMServerManager e2e_ppo_trainer_megatron_sglang #7060: Pull request #5430 synchronize by wuxibin89
26m 16s wuxibin89:wuxibin/resume_client
wuxibin89:wuxibin/resume_client
26m 16s
[model] feat: add qwen3.5 fsdp grpo training support with npu optimization e2e_ppo_trainer_megatron_sglang #7059: Pull request #5434 opened by Kirrito-k423
Action required Kirrito-k423:qwen3.5_fsdp_npu
Kirrito-k423:qwen3.5_fsdp_npu
Action required
[BREAKING][rollout,cfg] refactor: get rid of actor_rollout_ref config… e2e_ppo_trainer_megatron_sglang #7058: Commit de6a1bc pushed by ArronHZG
27m 20s main
main
27m 20s
[megatron, vllm] feat: NVFP4 (W4A16) QAT training support via ModelOpt e2e_ppo_trainer_megatron_sglang #7057: Pull request #5254 synchronize by jQizhang
Action required jQizhang:nvfp4_cleanup
jQizhang:nvfp4_cleanup
Action required
[WIP][algo] Migrate and implement the GDPO algorithm into the existing framework. e2e_ppo_trainer_megatron_sglang #7056: Pull request #5422 synchronize by Rhetee
Action required Rhetee:GDPO
Rhetee:GDPO
Action required
[WIP][algo] Migrate and implement the GDPO algorithm into the existing framework. e2e_ppo_trainer_megatron_sglang #7055: Pull request #5422 synchronize by Rhetee
Action required Rhetee:GDPO
Rhetee:GDPO
Action required
[WIP][algo] Migrate and implement the GDPO algorithm into the existing framework. e2e_ppo_trainer_megatron_sglang #7054: Pull request #5422 synchronize by Rhetee
Action required Rhetee:GDPO
Rhetee:GDPO
Action required
[rollout] refactor: bucketed transfer utils e2e_ppo_trainer_megatron_sglang #7053: Pull request #5309 synchronize by pengwu22
19m 24s pengwu22:pw/test-wt-utils
pengwu22:pw/test-wt-utils
19m 24s
[rollout] refactor: bucketed transfer utils e2e_ppo_trainer_megatron_sglang #7052: Pull request #5309 synchronize by pengwu22
35m 58s pengwu22:pw/test-wt-utils
pengwu22:pw/test-wt-utils
35m 58s
[rollout] feat: support auto resume on abort in FullyAsyncLLMServerManager e2e_ppo_trainer_megatron_sglang #7051: Pull request #5430 opened by wuxibin89
15m 23s wuxibin89:wuxibin/resume_client
wuxibin89:wuxibin/resume_client
15m 23s
[misc] feat: Add code for data grouping in no-padding scenario (#5424) e2e_ppo_trainer_megatron_sglang #7050: Commit c3fc222 pushed by vermouth1992
22m 6s main
main
22m 6s
[ci] feat: add profiling tests to vLLM ci e2e_ppo_trainer_megatron_sglang #7049: Pull request #5215 synchronize by Gary-cjy
Action required Gary-cjy:main
Gary-cjy:main
Action required
[BREAKING][rollout,cfg] refactor: get rid of actor_rollout_ref config from rollout e2e_ppo_trainer_megatron_sglang #7048: Pull request #5418 synchronize by wuxibin89
53m 32s wuxibin89:wuxibin/refactor_config
wuxibin89:wuxibin/refactor_config
53m 32s
[tool] refactor: remove tool schema plumbing from SingleTurnAgentLoop… e2e_ppo_trainer_megatron_sglang #7047: Commit 48b367b pushed by wuxibin89
1h 8m 14s main
main
1h 8m 14s
[megatron, vllm] feat: NVFP4 (W4A16) QAT training support via ModelOpt e2e_ppo_trainer_megatron_sglang #7046: Pull request #5254 synchronize by jQizhang
Action required jQizhang:nvfp4_cleanup
jQizhang:nvfp4_cleanup
Action required
[perf, trtllm] feat: Add Nsight support for rollout server mode (trtllm) e2e_ppo_trainer_megatron_sglang #7045: Pull request #5391 synchronize by davidmlw
40m 8s joyang-nv:liweim/nsys
joyang-nv:liweim/nsys
40m 8s
[trainer] feat: Add Nemo-Automodel as alternative training engine e2e_ppo_trainer_megatron_sglang #7044: Pull request #5407 synchronize by HuiyingLi
Queued HuiyingLi:add_automodel_sft_backend
HuiyingLi:add_automodel_sft_backend
Queued
[tool, rollout] refactor: cache tool list initialization with MCP timeout e2e_ppo_trainer_megatron_sglang #7043: Pull request #5426 opened by denismegerle
Action required denismegerle:public/tool-list-caching
denismegerle:public/tool-list-caching
Action required
[trainer] feat: Add Nemo-Automodel as alternative training engine e2e_ppo_trainer_megatron_sglang #7042: Pull request #5407 synchronize by HuiyingLi
Action required HuiyingLi:add_automodel_sft_backend
HuiyingLi:add_automodel_sft_backend
Action required
[trainer] feat: Add Nemo-Automodel as alternative training engine e2e_ppo_trainer_megatron_sglang #7041: Pull request #5407 synchronize by HuiyingLi
Action required HuiyingLi:add_automodel_sft_backend
HuiyingLi:add_automodel_sft_backend
Action required
[WIP][algo] Migrate and implement the GDPO algorithm into the existing framework. e2e_ppo_trainer_megatron_sglang #7040: Pull request #5422 synchronize by Rhetee
Action required Rhetee:GDPO
Rhetee:GDPO
Action required
[tool] refactor: remove tool schema plumbing from SingleTurnAgentLoop e2e_ppo_trainer_megatron_sglang #7039: Pull request #5425 opened by denismegerle
1h 13m 12s denismegerle:public/single-turn-disable-tools
denismegerle:public/single-turn-disable-tools
1h 13m 12s
[trainer] feat: add support for the GDPO algorithm e2e_ppo_trainer_megatron_sglang #7038: Pull request #5409 synchronize by yue-zeng-yue
Action required yue-zeng-yue:feat-gdpo
yue-zeng-yue:feat-gdpo
Action required
[rollout, data] fix: honor train_max_samples/val_max_samples in fully… e2e_ppo_trainer_megatron_sglang #7037: Commit c3e3970 pushed by ArronHZG
1h 50m 51s main
main
1h 50m 51s