Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Coding benchmark
SWE-bench Multimodal. Every model the catalog carries a published SWE Multimodal value for, ranked by that value.
A multimodal variant of SWE-bench that adds visual context such as screenshots and design mockups to software engineering issue descriptions. 7 models in the ModelChoice catalog have a published value.
Each point is a model with a published SWE Multimodal value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
7 models with a published SWE Multimodal value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Code patch generation with visual context |
|---|---|---|---|
| 1 | Anthropic | 74.1 | |
| 2 | Anthropic | 61.4 | |
| 3 | Anthropic | 59.4 |
What the benchmark measures and how its value is published, from the source’s own record.
SWE Multimodal (SWE-bench Multimodal) is a coding benchmark published by SWE-bench Multimodal. A multimodal variant of SWE-bench that adds visual context such as screenshots and design mockups to software engineering issue descriptions. Its evaluation set is described by the source as multimodal software engineering tasks. The source rates its difficulty as frontier multimodal coding.
Each model's result is published by SWE-bench Multimodal as code patch generation with visual context. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 7 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| 4 | Anthropic | 54.7 |
| 5 | Anthropic | 54.3 |
| 6 | Anthropic | 38.4 |
| 7 | Anthropic | 28.1 |
Evidence key: Observed
Rows are ordered by the value SWE-bench Multimodal published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.