Skip to content

Actions: verl-project/verl

Actions

reward_model_sglang

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
reward_model_sglang.yml
3,132 workflow runs
3,132 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[fsdp,vllm,trainer,algo] feat: On-Policy Distillation reward_model_sglang #4103: Pull request #4897 synchronize by JacobHelwig
2m 43s JacobHelwig:jhelwig/onPolicyDistillation
JacobHelwig:jhelwig/onPolicyDistillation
2m 43s
[BREAKING][rollout,cfg] refactor: get rid of actor_rollout_ref config from rollout reward_model_sglang #4102: Pull request #5418 opened by wuxibin89
4m 36s wuxibin89:wuxibin/refactor_config
wuxibin89:wuxibin/refactor_config
4m 36s
[fsdp,vllm,trainer,algo] feat: On-Policy Distillation reward_model_sglang #4101: Pull request #4897 synchronize by JacobHelwig
7m 15s JacobHelwig:jhelwig/onPolicyDistillation
JacobHelwig:jhelwig/onPolicyDistillation
7m 15s
[fsdp,vllm,trainer,algo] feat: On-Policy Distillation reward_model_sglang #4100: Pull request #4897 synchronize by JacobHelwig
7m 18s JacobHelwig:jhelwig/onPolicyDistillation
JacobHelwig:jhelwig/onPolicyDistillation
7m 18s
[fsdp,vllm,trainer,algo] feat: On-Policy Distillation reward_model_sglang #4099: Pull request #4897 synchronize by JacobHelwig
7m 13s JacobHelwig:jhelwig/onPolicyDistillation
JacobHelwig:jhelwig/onPolicyDistillation
7m 13s
[fsdp,vllm,trainer,algo] feat: On-Policy Distillation reward_model_sglang #4098: Pull request #4897 synchronize by JacobHelwig
26m 35s JacobHelwig:jhelwig/onPolicyDistillation
JacobHelwig:jhelwig/onPolicyDistillation
26m 35s
[megatron] feat: enhance model offloading and loading for frozen para… reward_model_sglang #4097: Commit b5979db pushed by vermouth1992
32m 49s main
main
32m 49s
[megatron] fix: missing model offload to CPU for forward_only mode (#… reward_model_sglang #4096: Commit 6b0bff3 pushed by vermouth1992
8m 45s main
main
8m 45s
[algo] fix: seq mean and default scale factor loss_mask.shape[-1] a… reward_model_sglang #4094: Commit b8d91ef pushed by tongyx361
8m 24s main
main
8m 24s
[trainer] feat: add padding for tensor alignment in preprocess_thd_no_padding function reward_model_sglang #4093: Pull request #5410 synchronize by RobotGF
18m 34s RobotGF:fix_mcore_cp
RobotGF:fix_mcore_cp
18m 34s
[megatron] feat: enhance model offloading and loading for frozen parameters reward_model_sglang #4092: Pull request #5412 synchronize by RobotGF
22m 28s RobotGF:fix_lora_offload
RobotGF:fix_lora_offload
22m 28s
[ckpt] feat: implement large tensor slicing in vllm rollout and CheckpointEngine for weight updating reward_model_sglang #4091: Pull request #5378 synchronize by jianjunzhong
9m 5s jianjunzhong:feat/chunked_weight_update
jianjunzhong:feat/chunked_weight_update
9m 5s
[algo] fix: seq mean and default scale factor loss_mask.shape[-1] as in seq-mean-token-sum-norm reward_model_sglang #4090: Pull request #5417 opened by tongyx361
54m 14s tongyx361:tyx/fix/seq-mean-in-seq-mean-token-sum-norm
tongyx361:tyx/fix/seq-mean-in-seq-mean-token-sum-norm
54m 14s
[tool] feature: scheduling analysis based on profiling data for torch profiler reward_model_sglang #4089: Pull request #5367 synchronize by Rhetee
Action required Rhetee:main
Rhetee:main
Action required
[doc] chore: npu best practice doc (#5415) reward_model_sglang #4088: Commit 03396b0 pushed by wucong25
1h 25m 43s main
main
1h 25m 43s
[perf, trtllm] feat: Add Nsight support for rollout server mode (trtllm) reward_model_sglang #4087: Pull request #5391 synchronize by davidmlw
1h 2m 10s joyang-nv:liweim/nsys
joyang-nv:liweim/nsys
1h 2m 10s
[Megatron] feat: Support routing replay on NPU with performance and compatibility enhancements reward_model_sglang #4086: Pull request #5298 synchronize by 755651978
1h 33m 12s 755651978:main-0212
755651978:main-0212
1h 33m 12s
[algo] feat: add DPPO with binary TV or binary KL implementation (#5397) reward_model_sglang #4085: Commit 182383b pushed by tongyx361
28m 58s main
main
28m 58s
[doc] chore: ascend update gspo optimization practice document (#5408) reward_model_sglang #4084: Commit 30b290e pushed by wucong25
6m 42s main
main
6m 42s
[Megatron] feat: Support routing replay on NPU with performance and compatibility enhancements reward_model_sglang #4083: Pull request #5298 synchronize by 755651978
24m 0s 755651978:main-0212
755651978:main-0212
24m 0s
[Megatron] feat: Support routing replay on NPU with performance and compatibility enhancements reward_model_sglang #4081: Pull request #5298 synchronize by 755651978
8m 45s 755651978:main-0212
755651978:main-0212
8m 45s
[fsdp,algo] feat: Support QAT (NVFP4) in FSDPEngine for the unified engine_workers architecture reward_model_sglang #4080: Pull request #5411 synchronize by zhangyimi
Action required zhangyimi:qat-core-v2
zhangyimi:qat-core-v2
Action required
[fsdp,algo] feat: Support QAT (NVFP4) in FSDPEngine for the unified engine_workers architecture reward_model_sglang #4079: Pull request #5411 synchronize by zhangyimi
Action required zhangyimi:qat-core-v2
zhangyimi:qat-core-v2
Action required
[fsdp,algo] feat: Support QAT (NVFP4) in FSDPEngine for the unified engine_workers architecture reward_model_sglang #4078: Pull request #5411 opened by zhangyimi
Action required zhangyimi:qat-core-v2
zhangyimi:qat-core-v2
Action required
[misc,trainer,rollout] feat: add Prometheus metrics logging to experiment tracking reward_model_sglang #4076: Pull request #5291 synchronize by guillemgt
Action required guillemgt:guillem.tarrach/upstream-prometheus-metrics
guillemgt:guillem.tarrach/upstream-prometheus-metrics
Action required