Skip to main content

SimpleVQA

A visual question answering benchmark focused on straightforward image-grounded understanding.

Benchmark score on SimpleVQA — June 2, 2026

BenchLM mirrors the published score view for SimpleVQA. Step 3.7 Flash leads the public snapshot at 79.2% , followed by Gemini 3.1 Pro (72.4%) and Muse Spark (71.3%). BenchLM does not use these results to rank models overall.

7 modelsMultimodal & GroundedCurrentDisplay onlyUpdated June 2, 2026

The published SimpleVQA snapshot is tightly clustered at the top: Step 3.7 Flash sits at 79.2%, while the third row is only 7.9 points behind. The broader top-10 spread is 23.1 points, so the benchmark still separates strong models even when the leaders cluster.

7 models have been evaluated on SimpleVQA. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. SimpleVQA is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About SimpleVQA

Year

2026

Tasks

Visual QA tasks

Format

Image-grounded question answering

Difficulty

General visual understanding

BenchLM uses SimpleVQA as a display-only visual QA reference rather than a weighted multimodal ranking input.

BenchLM freshness & provenance

Version

SimpleVQA 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

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.

Benchmark score table (7 models)

1
79.2%
2
72.4%
3
71.3%
4
61.1%
5
58.9%
6
57.4%
7
56.1%

FAQ

What does SimpleVQA measure?

A visual question answering benchmark focused on straightforward image-grounded understanding.

Which model scores highest on SimpleVQA?

Step 3.7 Flash by StepFun currently leads with a score of 79.2% on SimpleVQA.

How many models are evaluated on SimpleVQA?

7 AI models have been evaluated on SimpleVQA on BenchLM.

Last updated: June 2, 2026 · BenchLM version SimpleVQA 2026

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.