Benchmark · Reasoning and knowledge
SimpleQA Verified
1,000 short factual questions with no answer choices, made by Google from OpenAI's SimpleQA. Epoch AI asks the model to guess rather than abstain, has a grader model mark each answer and reports the share answered correctly, not Google's F1 score.
- Score
- % of questions correct, higher is better
- Run by
- Epoch AI
- Status
- Current, newest result 19 November 2026
- Updated
- as models are released
- Data
- the published file, CC BY 4.0, read 20 November 2026
Results
30 models we list, each at its best setting, best first.
| # | Model | Score | Range | Who ran it | Setting | Date | |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.2 A | 94.0% | Independent | effort max | 2026-11-06 | source | |
| 2 | GPT-6.2 Astra AA | 78.5% | Independent | effort max | 2026-10-27 | source | |
| 3 | Gemini 4 Carbon B | 77.7% | Independent | effort high | 2026-11-02 | source | |
| 4 | Gemini 4 Argon B | 76.0% | Independent | effort high | 2026-10-13 | source | |
| 5 | GPT-6 Astra A | 75.6% | Independent | effort max | 2026-08-30 | source | |
| 6 | Gemini 4 Neon B | 75.2% | Independent | effort xhigh | 2026-11-19 | source | |
| 7 | GPT-6.1 Sol AA | 73.9% | Independent | effort max | 2026-09-29 | source | |
| 8 | Gemini 3.1 Pro (preview) AA | 73.5% | Independent | effort high | 2026-08-10 | source | |
| 9 | Claude Opus 5.5 A | 72.2% | Independent | effort max | 2026-09-22 | source | |
| 10 | Gemini 4 Hydrogen B | 70.9% | Independent | effort xhigh | 2026-11-13 | source | |
| 11 | Claude Fable 5.1 A | 70.8% | Independent | effort max | 2026-09-01 | source | |
| 12 | Gemini 3.8 Flash AA | 69.7% | Independent | effort high | 2026-09-02 | source | |
| 13 | Kimi K3.1 Pro D | 62.2% | Independent | effort max | 2026-11-06 | source | |
| 14 | GPT-6 Sol A | 60.7% | Independent | effort max | 2026-09-22 | source | |
| 15 | GPT-6.1 Terra AA | 57.3% | Independent | effort max | 2026-10-26 | source | |
| 16 | Muse Spark Open C | 56.6% | Independent | effort xhigh | 2026-10-15 | source | |
| 17 | Grok 4.7 BB | 56.0% | Independent | effort xhigh | 2026-09-22 | source | |
| 18 | Kimi K3 B | 50.6% | Independent | effort max | 2026-08-27 | source | |
| 19 | DeepSeek V4 Pro D | 47.0% | Independent | effort max | 2026-08-27 | source | |
| 20 | Claude Sonnet 5.5 A | 46.5% | Independent | effort max | 2026-09-29 | source | |
| 21 | DeepSeek V4.1 Flash C | 42.9% | Independent | effort max | 2026-10-13 | source | |
| 22 | GPT-6 Luna A | 41.4% | Independent | effort max | 2026-09-22 | source | |
| 23 | GLM-5.3 B | 41.0% | Independent | effort max | 2026-08-28 | source | |
| 24 | Inkling B | 40.3% | Independent | effort xhigh | 2026-08-27 | source | |
| 25 | DeepSeek-V4-Flash-0731 C | 33.6% | Independent | effort max | 2026-08-27 | source | |
| 26 | Grok 4.3 BB | 33.2% | Independent | effort high | 2026-08-27 | source | |
| 27 | Claude Haiku 5.5 A | 23.8% | Independent | effort max | 2026-10-09 | source | |
| 28 | Mistral Large 4 C | 22.4% | Independent | effort high | 2026-11-02 | source | |
| 29 | Inkling-Small B | 19.1% | Independent | effort xhigh | 2026-08-27 | source | |
| 30 | Gemma 4 31B IT BB | 10.4% | Independent | 2026-08-27 | source |
Every setting, 31 rows
| Model | As the leaderboard names it | Score | Who ran it | Setting | Date | |
|---|---|---|---|---|---|---|
| Claude Fable 5.2 | claude-fable-5-2 | 94.0% | Independent | effort max | 2026-11-06 | source |
| GPT-6.2 Astra | gpt-6.2-astra | 78.5% | Independent | effort max | 2026-10-27 | source |
| Gemini 4 Carbon | gemini-4-carbon | 77.7% | Independent | effort high | 2026-11-02 | source |
| Gemini 4 Argon | gemini-4-argon | 76.0% | Independent | effort high | 2026-10-13 | source |
| GPT-6 Astra | gpt-6-astra | 75.6% | Independent | effort max | 2026-08-30 | source |
| Gemini 4 Neon | gemini-4-neon | 75.2% | Independent | effort xhigh | 2026-11-19 | source |
| GPT-6.1 Sol | gpt-6.1-sol | 73.9% | Independent | effort max | 2026-09-29 | source |
| Gemini 3.1 Pro (preview) | gemini-3.1-pro-preview | 73.5% | Independent | effort high | 2026-08-10 | source |
| Claude Opus 5.5 | claude-opus-5-5 | 72.2% | Independent | effort max | 2026-09-22 | source |
| Gemini 4 Hydrogen | gemini-4-hydrogen | 70.9% | Independent | effort xhigh | 2026-11-13 | source |
| Claude Fable 5.1 | claude-fable-5-1 | 70.8% | Independent | effort max | 2026-09-01 | source |
| Gemini 3.8 Flash | gemini-3.8-flash | 69.7% | Independent | effort high | 2026-09-02 | source |
| Kimi K3.1 Pro | kimi-k3.1-pro-together | 62.2% | Independent | effort max | 2026-11-06 | source |
| GPT-6 Sol | gpt-6-sol | 60.7% | Independent | effort max | 2026-09-22 | source |
| GPT-6.1 Terra | gpt-6.1-terra | 57.3% | Independent | effort max | 2026-10-26 | source |
| Muse Spark Open | Muse-Spark-Open-816B-A42B | 56.6% | Independent | effort xhigh | 2026-10-15 | source |
| Grok 4.7 | grok-4.7 | 56.0% | Independent | effort xhigh | 2026-09-22 | source |
| Kimi K3 | kimi-k3 | 50.6% | Independent | effort max | 2026-08-27 | source |
| DeepSeek V4 Pro | deepseek-v4-pro | 47.0% | Independent | effort max | 2026-08-27 | source |
| Claude Sonnet 5.5 | claude-sonnet-5-5 | 46.5% | Independent | effort max | 2026-09-29 | source |
| DeepSeek V4.1 Flash | deepseek-v4.1-flash | 42.9% | Independent | effort max | 2026-10-13 | source |
| DeepSeek V4.1 Flash | deepseek-v4.1-flash-1017 | 42.5% | Independent | effort max | 2026-10-19 | source |
| GPT-6 Luna | gpt-6-luna | 41.4% | Independent | effort max | 2026-09-22 | source |
| GLM-5.3 | glm-5.3 | 41.0% | Independent | effort max | 2026-08-28 | source |
| Inkling | Inkling | 40.3% | Independent | effort xhigh | 2026-08-27 | source |
| DeepSeek-V4-Flash-0731 | deepseek-v4-flash-0731 | 33.6% | Independent | effort max | 2026-08-27 | source |
| Grok 4.3 | grok-4.3 | 33.2% | Independent | effort high | 2026-08-27 | source |
| Claude Haiku 5.5 | claude-haiku-5-5 | 23.8% | Independent | effort max | 2026-10-09 | source |
| Mistral Large 4 | mistral-large-4-0-oss | 22.4% | Independent | effort high | 2026-11-02 | source |
| Inkling-Small | inkling-small | 19.1% | Independent | effort xhigh | 2026-08-27 | source |
| Gemma 4 31B IT | gemma-4-31b-it | 10.4% | Independent | 2026-08-27 | source |
A rank is among the models we list with a result, not a place on the whole leaderboard, and a model offered by several hosts is shown once, on its own listing. These results don't move our grades. Data: Epoch AI, Capabilities and benchmarking (CC BY 4.0), CC BY 4.0.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.