SimpleVQA

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

In index
Multimodal
Best in multimodal

Models18

Top result81.7%Qwen3.7 Plus

Top-3 spread6.7 pts

Year—

Result and price

13 of 18 results

Results

18 results

#ModelResult

01Qwen3.7 Plus81.7%

02Step 3.7 Flash79.2%

03Qwen3.8 Max75.0%

04Muse Spark Open73.6%

05Gemini 3 Pro73.2%

06dots3-note Preview72.5%

07Gemini 3.1 Pro72.4%

08Muse Spark71.3%

09Kimi K2.571.2%

10Qwen3.5 397B67.1%

11Claude Opus 4.565.7%

12GPT-5.461.1%

13Qwen3.6-35B-A3B58.9%

14Grok 4.2057.4%

15Qwen3.6-27B56.1%

16GPT-5.255.8%

17LFM3-VL-2B36.3%

18LFM2.5-VL-3B35.4%