AF_RL / discrete-firsterror-20260817

Commit History

Fix load example: Qwen3-VL needs AutoModelForImageTextToText, not AutoModelForCausalLM
6cc8fb3
verified

YeMoKoo commited on

Add first_error GRPO LoRA adapters (scalar / discrete), 2026-08-17
fb502e6
verified

YeMoKoo commited on