---
license: apache-2.0
base_model: Qwen/Qwen3.5-397B-A17B
library_name: transformers
pipeline_tag: image-text-to-text
tags:
- skyrl
- reinforcement-learning
---

# Qwen3.5-397B-A17B-Mercor

RL post-trained **Qwen3.5-397B-A17B** on APEX-Agents off-the-shelf datasets.

For more details, see the blog post:
[**Training frontier knowledge work agents: a 397B RL training guide with SkyRL**](https://www.mercor.com/blog/training-frontier-knowledge-work-agents-a-397b-rl-training-guide-with-skyrl)