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.
Multimodal & Grounded benchmark
Every model the catalog carries a published SimpleVQA value for, ranked by that value.
A visual question answering benchmark focused on straightforward image-grounded understanding. 18 models in the ModelChoice catalog have a published value.
Each point is a model with a published SimpleVQA value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
18 models with a published SimpleVQA 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 | Image-grounded question answering |
|---|---|---|---|
| 1 | Alibaba | 81.7 | |
| 2 | StepFun | 79.2 | |
| 3 | Alibaba | 75.0 |
What the benchmark measures and how its value is published, from the source’s own record.
SimpleVQA is a multimodal & grounded benchmark published by GLM-5V-Turbo. A visual question answering benchmark focused on straightforward image-grounded understanding. Its evaluation set is described by the source as visual QA tasks. The source rates its difficulty as general visual understanding.
Each model's result is published by GLM-5V-Turbo as image-grounded question answering. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 18 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.
All leaderboards · Multimodal & Grounded capability leaderboard
| 4 | Meta | 73.6 |
| 5 | 73.2 |
| 6 | DSdots3-note Preview | Dots Studio | 72.5 |
| 7 | 72.4 |
| 8 | Meta | 71.3 |
| 9 | Moonshot AI | 71.2 |
| 10 | Alibaba | 67.1 |
| 11 | Anthropic | 65.7 |
| 12 | OpenAI | 61.1 |
| 13 | Alibaba | 58.9 |
| 14 | xAI | 57.4 |
| 15 | Alibaba | 56.1 |
| 16 | OpenAI | 55.8 |
| 17 | LiquidAI | 36.3 |
| 18 | LiquidAI | 35.4 |
Evidence key: Observed
Rows are ordered by the value GLM-5V-Turbo 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.