Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and no NCCL
Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and no NCCL AsyncGRPOTrainer can now train a LoRA adapter and sync only that adapter to vLLM (TRL v1.14). A rank-1 adapter is a few megabytes, so it can travel through a Storage...
Read more: https://huggingface.co/blog/asyncgrpo-lora-hfjobs
Scan with your phone
Scan with your phone