CoderForge-Preview: SOTA open dataset for training efficient coding agents
Together AI Blog
•
Machine Learning
Open Source AI
We release CoderForge-Preview - the largest open test-verified coding agent dataset. By leveraging it to fine-tune Qwen-3 32B, we boost SWE-Bench Verified performance 23.0% above the base model reaching 59.4%, ranking among open-data models in the ≤32B parameter range.