Benchmark

SWE-Bench Multimodal

SWE-Bench Multimodal extends SWE-Bench to evaluate language models on software engineering tasks that involve visual inputs such as screenshots, UI…

Modality
multimodal
Categories
multimodal, reasoning, agents, code, vision
Openness
unknown
Source
llm_stats
Reported scores
4

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Claude Mythos PreviewAnthropic0.592026-04-07evidence
Claude Opus 4.8Anthropic0.3842026-05-28evidence
Claude Sonnet 5Anthropic0.2812026-06-30evidence
Qwen3.8-27BQwen0.3862026-08-14evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.