You can deploy this model and start earning money today!

skylenage-ai/
SWE-CI

Views

No views yet

SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous Integration

🔗 Github Link: https://github.com/SKYLENAGE-AI/SWE-CI 🔗 Paper Link: https://arxiv.org/pdf/2603.03823

📢 News (2026/4/20): We have released SWE-CI version 2.0. Building on v1.0, we have introduced an additional 126 new tasks. We also provide a new split: default_v2. Like the original default split, default_v2 consists of 100 tasks — however, it is more challenging (Qwen3.6-plus score drops… See the full description on the dataset page: https://huggingface.co/datasets/skylenage-ai/SWE-CI.