LLM Reference
activeCodingVision

SWE-bench Multimodal

Metric: % Resolved (higher is better)Introduced: 2025

A 517-issue extension of SWE-bench for visual software repositories. Tasks require resolving real GitHub issues where screenshots, plots, diagrams, or other visual evidence are part of the software-engineering context; model scores must preserve the evaluator and harness. High benchmark score alone doesn't make a model the right pick — weigh it against pricing, API availability, and release date.

Models ranked

1

tracked on this benchmark

Score band

59.4 – 59.4

best → lowest tracked

Snapshot trend

need ≥2 snapshots

Leaderboard

Tracked models ranked by % Resolved (higher is better).

Compare candidates
#Model variant and provenanceScore
1
Claude Opus 5

Configuration: Claude Opus 5 standard configuration

Version: SWE-bench MultimodalHarness: Anthropic internal SWE-bench Multimodal harnessEvaluator: AnthropicObserved: Jul 24, 2026Confidence: confirmedSource

Notes: Vendor-reported. Source: p.149, Table 8.1.A and Section 9.3. Do not compare this internal-harness row as identical to third-party runs.

59.4

How to read this benchmark

This benchmark scores models where higher is better. Scores are useful for directional filtering and shortlisting — not for universal quality ranking. Prefer benchmarks closest to your workload, then validate the linked model pages for pricing, context window, and provider availability.

Trust this score when

  • There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
  • The model list covers the same version family you can actually deploy today.
  • Top candidates overlap with your required routing and feature requirements.

Be cautious when

  • There is only one benchmark snapshot or the dataset appears stale.
  • The benchmark metric direction is opposite of your decision objective.
  • The score difference between options is narrow and likely within implementation variance.

FAQ

What does the SWE-bench Multimodal benchmark measure?

A 517-issue extension of SWE-bench for visual software repositories. Tasks require resolving real GitHub issues where screenshots, plots, diagrams, or other visual evidence are part of the software-engineering context; model scores must preserve the evaluator and harness. On this page it lists 1 tracked model variant where higher is better.

Is a higher SWE-bench Multimodal score always better?

For this benchmark, higher is better. A high score helps you shortlist, but confirm pricing, context window, and provider availability on each model page before committing — the top scorer is not always the right pick for your workload or budget.

How current is this SWE-bench Multimodal data?

This benchmark was last reviewed on Jul 26, 2026. Re-check the linked model pages for the freshest provider and pricing detail.

Related benchmarks

Last reviewed: Jul 26, 2026

Resources