32B and 72B (plus MT checkpoints)
Qwen license (models); mixed release components for data/pipeline include CC-BY-4.0 and Apache-2.0
Open-weight agentic software-engineering model family positioned as agent-native mid-training for repository-level coding workflows (navigate/edit/test loops) rather than single-turn code generation. Official release materials emphasize a scalable training recipe that combines contextually-native PR trajectories and environmentally-native executable trajectories, then applies environment-native SFT. The model card claims state-of-the-art performance among open training recipes with agentic scaffolds at these model sizes, reporting SWE-Bench Verified Pass@1 of 56.1% (32B) and 58.5% (72B), evaluated with SWE-Agent at temperature 0, 128k context, and a 100-step budget; release notes also claim generalization gains on HumanEval/EvalPlus and GPQA/SciBench. The release includes model checkpoints, dataset artifacts, and pipeline components for reproducible agentic training workflows.