Skip to content
AI Atlas

SWE-bench Multimodal — cost vs performance

Best current row per canonical model in the group “resolved · board=Multimodal · system=SWE-agent” (2 models) against total parameters. The dashed line is the Pareto frontier: no model is both better and cheaper than a point on it.

XOutput priceInput priceParametersContextMemory (est.)ScaleloglinearBubblecontextparamsnone
Group

No model has both a score in this group and a value on this axis

Try another axis or comparability group.