SimpleQA Verified: leaderboard
A thousand short factual questions with a single verifiable answer, filtered by Google from OpenAI's SimpleQA to drop duplicates, balance topics and correct answer keys. Models answer from memory with no browsing or tools, and a grader model marks each reply correct, incorrect or not attempted. This board reports the share of questions answered correctly, higher is better. Source: arxiv.org/abs/2509.07968.
Metric: Accuracy (%). Source: epoch.ai. Saturation forecast: Around May 2027. 91 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 76 | #8 (Gemini 4 Argon) |
| 2 | GPT-6 Astra (Max) | 75.6 | #6 (GPT-6 Astra) |
| 3 | GPT-6.1 Sol (Max) | 73.9 | #7 (GPT-6.1 Sol) |
| 4 | Gemini 3.1 Pro (Preview) (High) | 73.5 | #58 (Gemini 3.1 Pro (Preview)) |
| 5 | Claude Opus 5.5 (Max) | 72.2 | #3 (Claude Opus 5.5) |
| 6 | Claude Fable 5.1 (Max) | 70.8 | #9 (Claude Fable 5.1) |
| 7 | Claude Fable 5 (xHigh) | 70.7 | #16 (Claude Fable 5) |
| 8 | GPT-5.6 Sol (Max) | 69.7 | #19 (GPT-5.6 Sol) |
| 9 | Gemini 3.8 Flash (High) | 69.7 | #21 (Gemini 3.8 Flash) |
| 10 | Gemini 3.7 Flash (High) | 69.2 | #23 (Gemini 3.7 Flash) |
| 11 | Gemini 3 Flash (Preview) (High) | 66.8 | #82 (Gemini 3 Flash (Preview)) |
| 12 | Gemini 3.5 Flash (High) | 66.2 | #60 (Gemini 3.5 Flash) |
| 13 | Gemini 3.6 Flash (High) | 66.2 | #51 (Gemini 3.6 Flash) |
| 14 | GPT-5.5 (xHigh) | 63 | #31 (GPT-5.5) |
| 15 | GPT-6 Sol (Max) | 60.7 | #14 (GPT-6 Sol) |
No result here: #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=simpleqa-verified · How It Works · Data refreshed daily, snapshot 2026-10-20.