SimpleVQA leaderboard
As of November 14, 2026, among the results listed here, Qwen3.7 Plus has the highest reported SimpleVQA score, 81.7, published by Qwen on May 21, 2026, ahead of StepFun Step 3.7 Flash at 79.2. 6 models have a SimpleVQA result in our records.
Scores as the model makers published them, not re-run by us; each row links to its announcement.
6 models ranked · 4 makers
| Rank | Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|---|
| 1 | Qwen3.7 Plus | 81.7 | Not published | Qwen | May 21, 2026 |
| 2 | Step 3.7 Flash | 79.2 | Not published | StepFun | May 28, 2026 (model release) |
| 3 | Qwen3.8 Max (0902) | 75 | Not published | Qwen | August 3, 2026 |
| 4 | Kimi K2.5 | 71.2 | Thinking | Moonshot AI | January 27, 2026 (model release) |
| 5 | Qwen3.6 Plus | 69.4 | Not published | Qwen | May 21, 2026 |
| 6 | T Inkling 2 Small | 68.2 | Not published | Thinking Machines | November 9, 2026 (model release) |
Other reported runs
| Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|
| Qwen3.7 Plus | 70.3 | Not published | Qwen | August 3, 2026 |
How much to trust these numbers
- Source
- Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
- Settings
- Settings differ between makers. Check the Setting column before comparing two scores.
- Spread
- 81.7 to 68.2 across 6 models.
- Freshness
- Newest result published November 9, 2026; oldest January 27, 2026.
What SimpleVQA measures
SimpleVQA asks models to answer short natural-language questions using multiple modalities, with scores reflecting the factual correctness of their answers. Source: SimpleVQA: Multimodal Factuality Evaluation