II.
Benchmark overview
Reference · livebenchmark:swe-bench-multimodal
SWE-bench Multimodal overview
Multimodal extension of SWE-bench: visual / GUI-driven software engineering tasks.
Attributes
displayName
SWE-bench Multimodal
homepageUrl
kind
full-stack
targetsKind
AgentVersion
description
Multimodal extension of SWE-bench: visual / GUI-driven software engineering tasks.
Outgoing edges
applies_to1
- domain:software-engineering·DomainSoftware Engineering
Incoming edges
bounds_subject1
- scope-boundary:swe-bench-multimodal.scope·ScopeBoundary