Request for AgentRL Multi-Task / Single-Task Model Weights #8
Description
Hi! Thank you for the excellent and highly impactful work on AgentRL — it’s a very impressive framework and we really appreciate you open-sourcing the code.
I want to ask whether it would be possible to release the AgentRL-trained model weights, for example the Multi-Task and Single-Task models based on Qwen2.5-14B-Instruct.I noticed in a previous GitHub issue that you mentioned the models would be released, but I haven’t been able to find them on Hugging Face yet.
Thank you very much for your time and consideration!