-
Notifications
You must be signed in to change notification settings - Fork 4.1k
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
[Bug] vLLM rollout DP>1 fails when a RolloutReplica spans multiple nodes
bugSomething isn't workingSomething isn't workingStatus: Open.#6856 In verl-project/verl;- Status: Open.#6854 In verl-project/verl;
GRPO training wtih lora using qwen2.5-7b-vl
bugSomething isn't workingSomething isn't workingStatus: Open.#6851 In verl-project/verl;[Bug] FlashInfer FP16 MoE produces corrupted rollouts with Qwen3-Coder-30B-A3B in multi-node colocate-async
bugSomething isn't workingSomething isn't workingStatus: Open.#6847 In verl-project/verl;process_validation_metrics()crashes on None-filled sparsereward_extra_infokeys (metrics emitted for only some samples)bugSomething isn't workingSomething isn't workinghelp wantedExtra attention is neededExtra attention is neededStatus: Open.#6830 In verl-project/verl;Qwen3.5-4b ppo_kl != 0
bugSomething isn't workingSomething isn't workingStatus: Open.#6829 In verl-project/verl;- Status: Open.#6827 In verl-project/verl;
使用verl v0.8.0训练GLM-4.1V-9B-Thinking报错
bugSomething isn't workingSomething isn't workingStatus: Open.#6814 In verl-project/verl;[opd, megatron] 单teacher 模式,发现训练后期,ACC 会降到0
bugSomething isn't workingSomething isn't workingStatus: Open.#6811 In verl-project/verl;[OPD] [megatron/fsdp] use forward_kl_topk method occurs oom
bugSomething isn't workingSomething isn't workingStatus: Open.#6810 In verl-project/verl;opd, fsdp, 910b3, 教师模型是Qwen3-235B,两机部署教师也会OOM
bugSomething isn't workingSomething isn't workingStatus: Open.#6792 In verl-project/verl;dynamic-cp split the batch into local_cp_size sub-batches bug
bugSomething isn't workingSomething isn't workingStatus: Open.#6786 In verl-project/verl;