SWE-bench Multimodal (SWE Multimodal)
We show this table for reference; we do not rank on it.
A multimodal variant of SWE-bench that adds visual context such as screenshots and design mockups to software engineering issue descriptions.
Benchmark score on SWE Multimodal — November 11, 2026
We compile the SWE Multimodal rows from provider self-reports. Claude Fable 5.2 leads the table at 74.1%, followed by Claude Opus 5.5 (61.4%) and Claude Opus 5 (59.4%). We do not use these results to rank models overall.
Claude Fable 5.2
Anthropic
Claude Opus 5.5
Anthropic
Claude Opus 5
Anthropic
7 modelsCodingCurrentDisplay onlyUpdated November 11, 2026
Benchmark score table (7 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Claude Fable 5.2Anthropic | 74.1% | Not reported | Closed |
| 2 | Claude Opus 5.5Anthropic | 61.4% | Not reported | Closed |
| 3 | Claude Opus 5Anthropic | 59.4% | Not reported | Closed |
| 4 | Claude Fable 5.1Anthropic | 54.7% | Not reported | Closed |
| 5 | Claude Sonnet 5.5Anthropic | 54.3% | Not reported | Closed |
| 6 | Claude Opus 4.8Anthropic | 38.4% | Not reported | Closed |
| 7 | Claude Sonnet 5Anthropic | 28.1% | Not reported | Closed |
Among the reported SWE Multimodal rows, Claude Fable 5.2 is first at 74.1%. The third row is 14.7 points behind. The broader top-10 range is 46.0 points, so the table still separates the published systems.
7 models have been evaluated on SWE Multimodal. The benchmark falls in the Coding category. SWE Multimodal is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About SWE Multimodal
Year
2025
Tasks
Multimodal software engineering tasks
Format
Code patch generation with visual context
Difficulty
Frontier multimodal coding
BenchLM stores provider-reported SWE-bench Multimodal values in the coding category when the model vendor reports the benchmark as part of a software-engineering capability suite.
Freshness and provenance
Version
SWE Multimodal 2025
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does SWE Multimodal measure?
A multimodal variant of SWE-bench that adds visual context such as screenshots and design mockups to software engineering issue descriptions.
Which model scores highest on SWE Multimodal?
Claude Fable 5.2 by Anthropic currently leads with a score of 74.1% on SWE Multimodal.
How many models are evaluated on SWE Multimodal?
7 AI models have published results on SWE Multimodal in the BenchLM catalog.
Compare top models on SWE Multimodal
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.