multimodal benchmark
SWE-Bench Multimodal Leaderboard
SWE-Bench Multimodal extends SWE-Bench to evaluate language models on software engineering tasks that involve visual inputs such as screenshots, UI mockups, and diagrams alongside code understanding.
Updated Aug 17, 2026
Models4
Model coverage4
MetricScore
EvidenceB
SWE-Bench Multimodal Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score59.0% | Percentile100.0% | Participants4 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score38.6% | Percentile66.7% | Participants4 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score38.4% | Percentile33.3% | Participants4 | EvidenceC | Evaluated |
| Rank04 | ModelAN | Score28.1% | Percentile0.0% | Participants4 | EvidenceC | Evaluated |
SWE-Bench Multimodal Highlights
The leading models and scores on this benchmark.
SWE-Bench Multimodal Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for SWE-Bench Multimodal
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis swe-bench multimodal AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ANClaude Mythos PreviewAnthropic
- Score
- 59.0%
Strengths
- Ranks #1 of 4 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures SWE-Bench Multimodal, not total model capability
- 02ACQwen3.8-27BAlibaba Cloud / Qwen Team
- Score
- 38.6%
Strengths
- Ranks #2 of 4 compared models
- 67th percentile on this benchmark
- C evidence result
Considerations
- This result measures SWE-Bench Multimodal, not total model capability
- 03ANClaude Opus 4.8Anthropic
- Score
- 38.4%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 46 tok/s via Anthropic
Strengths
- Ranks #3 of 4 compared models
- 33th percentile on this benchmark
- C evidence result
Considerations
- This result measures SWE-Bench Multimodal, not total model capability
- 04ANClaude Sonnet 5Anthropic
- Score
- 28.1%
- Price
- $2.0 input / $10 output per 1M tokens
- Speed
- Up to 11 tok/s via Anthropic
Strengths
- Ranks #4 of 4 compared models
- 0th percentile on this benchmark
- C evidence result
Considerations
- This result measures SWE-Bench Multimodal, not total model capability
Selection summary
Best AI Models for SWE-Bench Multimodal
Claude Mythos Preview currently leads SWE-Bench Multimodal with 59.0%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is SWE-Bench Multimodal?
What SWE-Bench Multimodal measures and how its scores work.
SWE-Bench Multimodal extends SWE-Bench to evaluate language models on software engineering tasks that involve visual inputs such as screenshots, UI mockups, and diagrams alongside code understanding.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- SWE-Bench Multimodal
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- swe-bench-multimodal|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about SWE-Bench Multimodal.
Which model scores highest on SWE-Bench Multimodal?
Claude Mythos Preview is currently ranked first with 59.0%.
What does SWE-Bench Multimodal measure?
SWE-Bench Multimodal extends SWE-Bench to evaluate language models on software engineering tasks that involve visual inputs such as screenshots, UI mockups, and diagrams alongside code understanding.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
4 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.