SWE-bench Multimodal
A software-engineering benchmark of 517 real GitHub issues from visual JavaScript/web projects that include visual context (screenshots, UI mockups, diagrams), measuring whether AI systems can resolve issues whose fixes are verified by the repository's tests.
What this benchmark measures
A software-engineering benchmark of 517 real GitHub issues from visual JavaScript/web projects that include visual context (screenshots, UI mockups, diagrams), measuring whether AI systems can resolve issues whose fixes are verified by the repository's tests.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is % resolved. It should be interpreted within SWE-bench Multimodal, not compared as part of a site-wide ranking.
Frequently asked
What is SWE-bench Multimodal?
A software-engineering benchmark of 517 real GitHub issues from visual JavaScript/web projects that include visual context (screenshots, UI mockups, diagrams), measuring whether AI systems can resolve issues whose fixes are verified by the repository's tests. It is a coding benchmark measured by % resolved.
What does % resolved mean on SWE-bench Multimodal?
SWE-bench Multimodal reports % resolved (%); higher is better. Scores are shown only within SWE-bench Multimodal and are never averaged with other benchmarks.
What is the top reported SWE-bench Multimodal score?
Claude Mythos Preview has the top reported score on SWE-bench Multimodal: 59.0% (% resolved).
Why do SWE-bench Multimodal scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. SWE-bench Multimodal scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".