SWE-Bench ProMax

SWE-Bench ProMax extends the original SWE-Bench with more complex, realistic software-engineering tasks, designed by the SWE-Bench team to benchmark AI coding agents against harder, real-world challenges.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.