SimpleQA Verified
1,000 short factual questions about politics, science and technology, art, sports, geography, music and more.
- Unit
- % correct
- Results
- 91 results80 Tokscale models matched, 82 in the source
- Last update
- Oct 15, 2026Newest run date
- Retrieved
- Oct 21, 2026Source data through Oct 15, 2026
Conditions. Run by Epoch AI at each reasoning effort.
Attribution. Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks (CC BY 4.0)
Results
One row per model: its best reported result. The rank is the position in this list, by score; the source's own ranking can differ.
Tokscale usage counts accounts that used the model in public Tokscale submissions, Oct 14 – Nov 12, 2026. It is shown next to the score and does not explain it.
| Rank | Model | Score (% correct) | Conditions | Date | Tokscale usage (30 days) | Source link |
|---|---|---|---|---|---|---|
| 1 | Gemini 4 ArgonListed as gemini-4-argon_high | 76.0%Standard error ±1.4 pts | High effort | Oct 13, 2026Run date | 82active users | Source for Gemini 4 Argon (opens in a new tab) |
| 2 | GPT-6 AstraListed as gpt-6-astra_max | 75.6%Standard error ±1.4 pts | Max effort | Aug 30, 2026Run date | 562active users | Source for GPT-6 Astra (opens in a new tab) |
| 3 | GPT-6.1 SolListed as gpt-6.1-sol_max | 73.9%Standard error ±1.4 pts | Max effort | Sep 29, 2026Run date | 659active users | Source for GPT-6.1 Sol (opens in a new tab) |
| 4 | Gemini 3.1 Pro PreviewListed as gemini-3.1-pro-preview_high | 73.5%Standard error ±1.4 pts | High effort | Aug 10, 2026Run date | 5active users | Source for Gemini 3.1 Pro Preview (opens in a new tab) |
| 5 | Claude Opus 5.5Listed as claude-opus-5-5_max | 72.2%Standard error ±1.4 pts | Max effort | Sep 22, 2026Run date | 709active users | Source for Claude Opus 5.5 (opens in a new tab) |
| 6 | Claude Fable 5.1Listed as claude-fable-5-1_max | 70.8%Standard error ±1.4 pts | Max effort | Sep 1, 2026Run date | 361active users | Source for Claude Fable 5.1 (opens in a new tab) |
| 7 | Claude Fable 5Listed as claude-fable-5_xhigh | 70.7%Standard error ±1.4 pts | XHigh effort | Aug 10, 2026Run date | 41active users | Source for Claude Fable 5 (opens in a new tab) |
| 8 | Gemini 3.8 FlashListed as gemini-3.8-flash_high | 69.7%Standard error ±1.5 pts | High effort | Sep 2, 2026Run date | 66active users | Source for Gemini 3.8 Flash (opens in a new tab) |
| 8 | GPT-5.6 SolListed as gpt-5.6-sol_max | 69.7%Standard error ±1.5 pts | Max effort | Aug 10, 2026Run date | 217active users | Source for GPT-5.6 Sol (opens in a new tab) |
| 10 | Gemini 3.7 FlashListed as gemini-3.7-flash_high | 69.2%Standard error ±1.5 pts | High effort | Aug 27, 2026Run date | 4active users | Source for Gemini 3.7 Flash (opens in a new tab) |
| 11 | Gemini 3 Flash PreviewListed as gemini-3-flash-preview_high | 66.8%Standard error ±1.5 pts | High effort | Aug 27, 2026Run date | 4active users | Source for Gemini 3 Flash Preview (opens in a new tab) |
| 12 | Gemini 3.5 FlashListed as gemini-3.5-flash_high | 66.2%Standard error ±1.5 pts | High effort | Aug 27, 2026Run date | 3active users | Source for Gemini 3.5 Flash (opens in a new tab) |
| 12 | Gemini 3.6 FlashListed as gemini-3.6-flash_high | 66.2%Standard error ±1.5 pts | High effort | Aug 27, 2026Run date | 3active users | Source for Gemini 3.6 Flash (opens in a new tab) |
| 14 | GPT-5.5Listed as gpt-5.5_xhigh | 63.0%Standard error ±1.5 pts | XHigh effort | Aug 27, 2026Run date | 30active users | Source for GPT-5.5 (opens in a new tab) |
| 15 | GPT-6 SolListed as gpt-6-sol_max | 60.7%Standard error ±1.5 pts | Max effort | Sep 22, 2026Run date | 245active users | Source for GPT-6 Sol (opens in a new tab) |
| 16 | Muse Spark 1.2Listed as muse-spark-1.2_xhigh | 60.3%Standard error ±1.6 pts | XHigh effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for Muse Spark 1.2 (opens in a new tab) |
| 17 | Claude Opus 5Listed as claude-opus-5_max | 59.9%Standard error ±1.6 pts | Max effort | Aug 10, 2026Run date | 163active users | Source for Claude Opus 5 (opens in a new tab) |
| 18 | Muse Spark 1.1Listed as muse-spark-1.1 | 57.8%Standard error ±1.6 pts | Not reported | Aug 31, 2026Run date | No Tokscale usage in the last 30 days | Source for Muse Spark 1.1 (opens in a new tab) |
| 19 | Muse Spark OpenListed as Muse-Spark-Open-816B-A42B_xhigh | 56.6%Standard error ±1.6 pts | XHigh effort | Oct 15, 2026Run date | 2active users | Source for Muse Spark Open (opens in a new tab) |
| 20 | Grok 4.7Listed as grok-4.7_xhigh | 56.0%Standard error ±1.6 pts | XHigh effort | Sep 22, 2026Run date | 131active users | Source for Grok 4.7 (opens in a new tab) |
| 21 | Qwen3.7 MaxListed as qwen3.7-max | 55.8%Standard error ±1.6 pts | Not reported | Aug 27, 2026Run date | 3active users | Source for Qwen3.7 Max (opens in a new tab) |
| 22 | Claude Opus 4.8Listed as claude-opus-4-8_max | 53.0%Standard error ±1.6 pts | Max effort | Aug 27, 2026Run date | 222active users | Source for Claude Opus 4.8 (opens in a new tab) |
| 23 | DeepSeek V4 Pro 0813Listed as deepseek-v4-pro-0813_max | 52.9%Standard error ±1.6 pts | Max effort | Aug 27, 2026Run date | 6active users | Source for DeepSeek V4 Pro 0813 (opens in a new tab) |
| 24 | Qwen3.6 Max PreviewListed as qwen3.6-max-preview | 52.0%Standard error ±1.6 pts | Not reported | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for Qwen3.6 Max Preview (opens in a new tab) |
| 25 | Claude Opus 4.7Listed as claude-opus-4-7_xhigh | 51.7%Standard error ±1.6 pts | XHigh effort | Aug 27, 2026Run date | 27active users | Source for Claude Opus 4.7 (opens in a new tab) |
| 26 | Kimi K3Listed as kimi-k3_max | 50.6%Standard error ±1.6 pts | Max effort | Aug 27, 2026Run date | 56active users | Source for Kimi K3 (opens in a new tab) |
| 27 | GPT-5 2025 08 07Listed as gpt-5-2025-08-07_high | 50.1%Standard error ±1.6 pts | High effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for GPT-5 2025 08 07 (opens in a new tab) |
| 28 | o3Listed as o3-2025-04-16_high | 49.4%Standard error ±1.6 pts | High effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for o3 (opens in a new tab) |
| 29 | Grok 4.6Listed as grok-4.6_high | 49.3%Standard error ±1.6 pts | High effort | Aug 27, 2026Run date | 41active users | Source for Grok 4.6 (opens in a new tab) |
| 30 | Qwen3 MaxListed as qwen3-max-2025-09-23 | 48.7%Standard error ±1.6 pts | Not reported | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for Qwen3 Max (opens in a new tab) |
| 31 | Grok 4.5Listed as grok-4.5_high | 48.3%Standard error ±1.6 pts | High effort | Aug 27, 2026Run date | 9active users | Source for Grok 4.5 (opens in a new tab) |
| 32 | GPT-5.1Listed as gpt-5.1-2025-11-13_high | 48.0%Standard error ±1.6 pts | High effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for GPT-5.1 (opens in a new tab) |
| 33 | Qwen3.8 Max 0902Listed as qwen3.8-max-0902_xhigh | 47.3%Standard error ±1.6 pts | XHigh effort | Sep 2, 2026Run date | 1active users | Source for Qwen3.8 Max 0902 (opens in a new tab) |
| 34 | Claude Opus 4.6Listed as claude-opus-4-6_max | 47.0%Standard error ±1.6 pts | Max effort | Aug 27, 2026Run date | 52active users | Source for Claude Opus 4.6 (opens in a new tab) |
| 35 | DeepSeek V4 ProListed as deepseek-v4-pro_max | 47.0%Standard error ±1.6 pts | Max effort | Aug 27, 2026Run date | 51active users | Source for DeepSeek V4 Pro (opens in a new tab) |
| 36 | Claude Sonnet 5.5Listed as claude-sonnet-5-5_max | 46.5%Standard error ±1.6 pts | Max effort | Sep 29, 2026Run date | 555active users | Source for Claude Sonnet 5.5 (opens in a new tab) |
| 37 | GPT-5.4 ProListed as gpt-5.4-pro-2026-03-05_xhigh | 46.3%Standard error ±1.6 pts | XHigh effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for GPT-5.4 Pro (opens in a new tab) |
| 38 | Qwen3.8 MaxListed as qwen3.8-max_xhigh | 45.8%Standard error ±1.6 pts | XHigh effort | Aug 27, 2026Run date | 16active users | Source for Qwen3.8 Max (opens in a new tab) |
| 39 | Claude Opus 4.5Listed as claude-opus-4-5-20251101_32K | 45.7%Standard error ±1.6 pts | 32K thinking budget | Aug 27, 2026Run date | 4active users | Source for Claude Opus 4.5 (opens in a new tab) |
| 40 | GPT-5.4Listed as gpt-5.4-2026-03-05_xhigh | 45.1%Standard error ±1.6 pts | XHigh effort | Aug 27, 2026Run date | 1active users | Source for GPT-5.4 (opens in a new tab) |
| 41 | Qwen3.6 PlusListed as qwen3.6-plus | 44.1%Standard error ±1.6 pts | Not reported | Aug 27, 2026Run date | 1active users | Source for Qwen3.6 Plus (opens in a new tab) |
| 42 | GPT-5.6 TerraListed as gpt-5.6-terra_max | 43.2%Standard error ±1.6 pts | Max effort | Aug 10, 2026Run date | 119active users | Source for GPT-5.6 Terra (opens in a new tab) |
| 43 | DeepSeek V4.1 FlashListed as deepseek-v4.1-flash_max | 42.9%Standard error ±1.6 pts | Max effort | Oct 13, 2026Run date | 225active users | Source for DeepSeek V4.1 Flash (opens in a new tab) |
| 44 | GPT-6 LunaListed as gpt-6-luna_max | 41.4%Standard error ±1.6 pts | Max effort | Sep 22, 2026Run date | 432active users | Source for GPT-6 Luna (opens in a new tab) |
| 45 | o1Listed as o1-2024-12-17_high | 41.1%Standard error ±1.6 pts | High effort | Aug 31, 2026Run date | No Tokscale usage in the last 30 days | Source for o1 (opens in a new tab) |
| 46 | GLM-5.3Listed as glm-5.3_max | 41.0%Standard error ±1.6 pts | Max effort | Aug 28, 2026Run date | 138active users | Source for GLM-5.3 (opens in a new tab) |
| 46 | GPT-5.6 LunaListed as gpt-5.6-luna_max | 41.0%Standard error ±1.6 pts | Max effort | Aug 10, 2026Run date | 158active users | Source for GPT-5.6 Luna (opens in a new tab) |
| 48 | Qwen3-235B-A22B-Thinking-2507Not matched to a Tokscale model | 40.4%Standard error ±1.6 pts | Not reported | Aug 27, 2026Run date | Not matched to a Tokscale model | Source for Qwen3-235B-A22B-Thinking-2507 (opens in a new tab) |
| 49 | InklingListed as Inkling_xhigh | 40.3%Standard error ±1.6 pts | XHigh effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for Inkling (opens in a new tab) |
| 50 | GPT-5.2Listed as gpt-5.2-2025-12-11_xhigh | 37.1%Standard error ±1.5 pts | XHigh effort | Aug 27, 2026Run date | No Tokscale usage in the last 30 days | Source for GPT-5.2 (opens in a new tab) |
Showing 50 of 82 models
Score by release date
Each dot is one model's best result on SimpleQA Verified, placed at the model's release date. The dashed line is the best score among models released up to that date.
- Used on Tokscale in the last 30 days
- No Tokscale usage in the last 30 days
- Best score so far
Gemini 4 Argon: 76.0% · released Sep 30, 2026 · High effort · 82 active users on Tokscale in the last 30 days
80 models, released Jul 18, 2024 to Oct 15, 2026. Highest: Gemini 4 Argon, 76.0%, released Sep 30, 2026. Use the left and right arrow keys to read each model in release order.0%20%40%60%80%
Show data
| Model | Released | Score | Tokscale users, 30 days |
|---|---|---|---|
| Muse Spark Open | Oct 15, 2026 | 56.6%XHigh effort | 2 |
| Claude Haiku 5.5 | Oct 7, 2026 | 23.8%Max effort | 477 |
| Mistral Large 4 | Oct 6, 2026 | 20.0%High effort | 4 |
| Gemini 4 Argon | Sep 30, 2026 | 76.0%High effort | 82 |
| GPT-6.1 Sol | Sep 29, 2026 | 73.9%Max effort | 659 |
| Claude Sonnet 5.5 | Sep 28, 2026 | 46.5%Max effort | 555 |
| GPT-6 Luna | Sep 22, 2026 | 41.4%Max effort | 432 |
| GPT-6 Sol | Sep 22, 2026 | 60.7%Max effort | 245 |
| Claude Opus 5.5 | Sep 22, 2026 | 72.2%Max effort | 709 |
| Grok 4.7 | Sep 21, 2026 | 56.0%XHigh effort | 131 |
| DeepSeek V4.1 Flash | Sep 9, 2026 | 42.9%Max effort | 225 |
| GPT-6 Astra | Sep 3, 2026 | 75.6%Max effort | 562 |
| Gemini 3.8 Flash | Sep 2, 2026 | 69.7%High effort | 66 |
| Qwen3.8 Max 0902 | Sep 1, 2026 | 47.3%XHigh effort | 1 |
| Claude Fable 5.1 | Sep 1, 2026 | 70.8%Max effort | 361 |
| GLM-5.3 | Aug 14, 2026 | 41.0%Max effort | 138 |
| DeepSeek V4 Pro 0813 | Aug 13, 2026 | 52.9%Max effort | 6 |
| Gemini 3.7 Flash | Aug 13, 2026 | 69.2%High effort | 4 |
| Grok 4.6 | Aug 12, 2026 | 49.3%High effort | 41 |
| Muse Spark 1.2 | Aug 5, 2026 | 60.3%XHigh effort | No Tokscale usage in the last 30 days |
| Qwen3.8 Max | Aug 2, 2026 | 45.8%XHigh effort | 16 |
| DeepSeek V4 Flash 0731 | Jul 31, 2026 | 33.6%Max effort | 18 |
| Claude Opus 5 | Jul 24, 2026 | 59.9%Max effort | 163 |
| Gemini 3.6 Flash | Jul 21, 2026 | 66.2%High effort | 3 |
| Kimi K3 | Jul 16, 2026 | 50.6%Max effort | 56 |
| Inkling Small | Jul 15, 2026 | 19.1%XHigh effort | No Tokscale usage in the last 30 days |
| Inkling | Jul 15, 2026 | 40.3%XHigh effort | No Tokscale usage in the last 30 days |
| GPT-5.6 Luna | Jul 9, 2026 | 41.0%Max effort | 158 |
| GPT-5.6 Terra | Jul 9, 2026 | 43.2%Max effort | 119 |
| Muse Spark 1.1 | Jul 9, 2026 | 57.8% | No Tokscale usage in the last 30 days |
| GPT-5.6 Sol | Jul 9, 2026 | 69.7%Max effort | 217 |
| Grok 4.5 | Jul 8, 2026 | 48.3%High effort | 9 |
| Claude Sonnet 5 | Jun 30, 2026 | 33.7%Max effort | 138 |
| GLM-5.2 | Jun 16, 2026 | 34.2%Max effort | 13 |
| Kimi K2.7 Code | Jun 12, 2026 | 36.5% | 17 |
| Claude Fable 5 | Jun 9, 2026 | 70.7%XHigh effort | 41 |
| Claude Opus 4.8 | May 28, 2026 | 53.0%Max effort | 222 |
| Qwen3.7 Max | May 19, 2026 | 55.8% | 3 |
| Gemini 3.5 Flash | May 19, 2026 | 66.2%High effort | 3 |
| Qwen3.6 Flash | Apr 27, 2026 | 15.9% | No Tokscale usage in the last 30 days |
| DeepSeek V4 Pro | Apr 24, 2026 | 47.0%Max effort | 51 |
| GPT-5.5 | Apr 23, 2026 | 63.0%XHigh effort | 30 |
| Kimi K2.6 | Apr 20, 2026 | 34.9% | 4 |
| Qwen3.6 Max Preview | Apr 20, 2026 | 52.0% | No Tokscale usage in the last 30 days |
| Grok 4.3 | Apr 17, 2026 | 33.2%High effort | No Tokscale usage in the last 30 days |
| Claude Opus 4.7 | Apr 16, 2026 | 51.7%XHigh effort | 27 |
| GLM-5.1 | Apr 7, 2026 | 34.0% | 5 |
| Gemma 4 31B IT | Apr 2, 2026 | 10.4% | 3 |
| Qwen3.6 Plus | Mar 31, 2026 | 44.1% | 1 |
| GPT-5.4 Nano | Mar 17, 2026 | 11.7%High effort | 4 |
| GPT-5.4 Mini | Mar 17, 2026 | 29.4%High effort | 1 |
| GPT-5.4 | Mar 5, 2026 | 45.1%XHigh effort | 1 |
| GPT-5.4 Pro | Mar 5, 2026 | 46.3%XHigh effort | No Tokscale usage in the last 30 days |
| Qwen3.5 Flash | Feb 25, 2026 | 20.3% | No Tokscale usage in the last 30 days |
| Gemini 3.1 Pro Preview | Feb 19, 2026 | 73.5%High effort | 5 |
| Grok 4.20 0309 Reasoning | Feb 17, 2026 | 30.2% | 1 |
| Claude Sonnet 4.6 | Feb 17, 2026 | 35.5%High effort | 55 |
| Qwen3.5 Plus | Feb 16, 2026 | 25.4% | No Tokscale usage in the last 30 days |
| Claude Opus 4.6 | Feb 5, 2026 | 47.0%Max effort | 52 |
| Kimi K2.5 | Jan 27, 2026 | 34.3% | No Tokscale usage in the last 30 days |
| GLM-4.7 | Dec 22, 2025 | 32.2% | 4 |
| Gemini 3 Flash Preview | Dec 17, 2025 | 66.8%High effort | 4 |
| GPT-5.2 | Dec 11, 2025 | 37.1%XHigh effort | No Tokscale usage in the last 30 days |
| Claude Opus 4.5 | Nov 24, 2025 | 45.7%32K thinking budget | 4 |
| GPT-5.1 | Nov 13, 2025 | 48.0%High effort | No Tokscale usage in the last 30 days |
| Claude Haiku 4.5 | Oct 15, 2025 | 13.2% | 102 |
| Claude Sonnet 4.5 | Sep 29, 2025 | 30.7%59K thinking budget | 23 |
| Qwen3 Max | Sep 24, 2025 | 48.7% | No Tokscale usage in the last 30 days |
| GPT-5 Nano | Aug 7, 2025 | 11.7%High effort | No Tokscale usage in the last 30 days |
| GPT-5 Mini | Aug 7, 2025 | 21.6%High effort | 1 |
| GPT-5 2025 08 07 | Aug 7, 2025 | 50.1%High effort | No Tokscale usage in the last 30 days |
| o4-mini | Apr 16, 2025 | 19.6%Low effort | No Tokscale usage in the last 30 days |
| o3 | Apr 16, 2025 | 49.4%High effort | No Tokscale usage in the last 30 days |
| GPT-4.1 Nano | Apr 14, 2025 | 6.0% | No Tokscale usage in the last 30 days |
| GPT-4.1 Mini 2025 04 14 | Apr 14, 2025 | 12.7% | 5 |
| GPT-4.1 2025 04 14 | Apr 14, 2025 | 31.1% | 5 |
| o3-mini | Jan 31, 2025 | 15.3%High effort | No Tokscale usage in the last 30 days |
| o1 | Dec 17, 2024 | 41.1%High effort | No Tokscale usage in the last 30 days |
| GPT-4o 2024 08 06 | Aug 6, 2024 | 26.0% | No Tokscale usage in the last 30 days |
| GPT-4o Mini | Jul 18, 2024 | 8.3% | 1 |
Method and matching
How results are run
- Run by Epoch AI at each reasoning effort.
Matching to Tokscale models
- 89 of 91 results are matched to 80 Tokscale models: 63 by exact id, 26 by normalized name.
- No match is marked as reviewed by hand yet.
- 2 results from 2 models are not matched. They stay in the table without a Tokscale link.
- Reasoning effort, thinking budget and agent are conditions of a result. They never change which model a result belongs to.
License and attribution
- CC BY 4.0 (opens in a new tab), retrieved Oct 21, 2026.
- Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks (CC BY 4.0)
Sources and attribution
Each score keeps its own source, unit and run conditions. Tokscale does not combine benchmarks into one score, and a missing score means the model was not evaluated, not that it scored zero.