SimpleQA — OpenAI factuality bench; folds 'SimpleQA'/'(Correct)' raws. Distinct from simpleqa_verified.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | DeepSeek-V3.2-Exp | DeepSeek | 97.1 | 1 | 2026-08-23 |
| 2 | Qwen3.6 27B | Alibaba | 95.7 | 2 | 2026-06-19 |
| 3 | Grok 4 Fast | SpaceXAI | 95 | 2 | 2026-08-23 |
| 4 | DeepSeek-V3.1 | DeepSeek | 93.4 | 3 | 2026-08-23 |
| 5 | DeepSeek-R1 | DeepSeek | 92.3 | 17 | 2026-08-23 |
| 6 | Qwen3.5 9B | Alibaba | 91.2 | 1 | 2026-05-17 |
| 7 | gpt-oss-20b | OpenAI | 85.4 | 1 | 2026-05-17 |
| 8 | Grok 3 | SpaceXAI | 82 | 1 | 2026-07-13 |
| 9 | ERNIE 5.0 Thinking Preview | Baidu | 75 | 1 | 2026-08-23 |
| 10 | Gemini 3 Pro | 72.1 | 1 | 2026-08-23 | |
| 11 | Gemini 3 Flash Preview | 68.7 | 1 | 2026-08-23 | |
| 12 | GPT-4.5 Preview | OpenAI | 62.5 | 1 | 2026-08-23 |
| 13 | DeepSeek-V4-Pro | DeepSeek | 57.9 | 1 | 2026-08-23 |
| 14 | Qwen3 VL 32B Reasoning | Alibaba | 55.4 | 1 | 2026-08-23 |
| 15 | Qwen3 235B A22B Instruct 2507 | Alibaba | 54.3 | 1 | 2026-08-23 |
| 16 | Gemini 2.5 Pro Preview 06.05 (32k think) | 54 | 3 | 2026-08-24 | |
| 17 | Qwen3 VL 235B A22B Instruct | Alibaba | 51.9 | 1 | 2026-08-23 |
| 18 | Gemini 2.5 Pro | 50.8 | 1 | 2026-08-23 | |
| 19 | Qwen3 VL Thinking (8B) | Alibaba | 49.6 | 1 | 2026-08-23 |
| 20 | o3 | OpenAI | 49.4 | 2 | 2026-06-05 |
| 21 | Qwen3 VL 4B Instruct | Alibaba | 48 | 1 | 2026-08-23 |
| 22 | o1 | OpenAI | 47 | 6 | 2026-08-23 |
| 23 | Qwen3 VL 235B A22B Reasoning | Alibaba | 44.4 | 1 | 2026-08-23 |
| 24 | Grok 3 Beta | SpaceXAI | 43.6 | 1 | 2026-07-13 |
| 25 | Gemini 3.1 Flash Lite Preview | 43.3 | 1 | 2026-08-23 | |
| 26 | GPT-4o | OpenAI | 39.4 | 9 | 2026-08-23 |
| 27 | Kimi K2 Base | Moonshot | 35.3 | 7 | 2026-08-23 |
| 28 | Step3.5 Flash Base | StepFun | 31.6 | 2 | 2026-06-05 |
| 29 | Kimi K2 Instruct | Moonshot | 31 | 4 | 2026-08-23 |
| 30 | R1-Zero | DeepSeek | 30.3 | 1 | 2026-05-03 |
| 31 | GLM-4.5-Base | Z.ai | 30 | 4 | 2026-06-05 |
| 32 | Claude 3.5 Sonnet | Anthropic | 28.4 | 7 | 2026-06-06 |
| 33 | DeepSeek-V3 | DeepSeek | 27.7 | 10 | 2026-08-23 |
| 34 | Qwen3 VL 30B A3B Instruct | Alibaba | 27 | 1 | 2026-08-23 |
| 35 | DeepSeek-V3.2-Exp-Base | DeepSeek | 27 | 4 | 2026-06-13 |
| 36 | Gemini 2.5 Flash | 26.9 | 1 | 2026-08-23 | |
| 37 | Gemini 2.0 Flash Exp | 26.6 | 1 | 2026-06-06 | |
| 38 | Hy3 preview-Base | Tencent | 26.5 | 2 | 2026-06-05 |
| 39 | DeepSeek-V3.1-Base | DeepSeek | 26.3 | 4 | 2026-06-13 |
| 40 | DeepSeek-V3-Base | DeepSeek | 26.1 | 2 | 2026-06-05 |
| 41 | Qwen3 VL 30B A3B Reasoning | Alibaba | 23.9 | 1 | 2026-08-23 |
| 42 | Mistral Large | Mistral | 23.8 | 1 | 2026-08-23 |
| 43 | Mistral Large 3 675B Base 2512 | Mistral | 23.8 | 1 | 2026-08-23 |
| 44 | MiniMax Text 01 | MiniMax | 23.7 | 1 | 2026-06-06 |
| 45 | Qwen3 30B A3B | Alibaba | 23.6 | 1 | 2026-06-15 |
| 46 | Gemini 1.5 Pro | 23.4 | 1 | 2026-06-06 | |
| 47 | Llama 3.1 Instruct 405B | Meta | 23.2 | 2 | 2026-06-06 |
| 48 | Claude 4 Opus | Anthropic | 22.8 | 2 | 2026-06-15 |
| 49 | Grok 3 Mini Beta | SpaceXAI | 21.7 | 1 | 2026-07-13 |
| 50 | Llama 3.3 70B Instruct | Meta | 20.9 | 1 | 2026-05-30 |
| 51 | MiMo V2 Flash Base | Xiaomi | 20.6 | 4 | 2026-06-13 |
| 52 | MiniMax M1 80K | MiniMax | 18.5 | 3 | 2026-08-23 |
| 53 | MiniMax M1 40K | MiniMax | 17.9 | 3 | 2026-08-23 |
| 54 | Claude Sonnet 4 | Anthropic | 15.9 | 2 | 2026-06-15 |
| 55 | o3-mini | OpenAI | 15 | 1 | 2026-08-23 |
| 56 | Qwen3 235B A22B | Alibaba | 13.2 | 4 | 2026-06-15 |
| 57 | Gemini 2.5 Flash Lite | 10.7 | 1 | 2026-08-23 | |
| 58 | Granite 4.1 30B Base | IBM | 10.5 | 2 | 2026-06-15 |
| 59 | Qwen2.5 Instruct 72B | Alibaba | 10.3 | 3 | 2026-06-06 |
| 60 | DeepSeek-V2.5 | DeepSeek | 10.2 | 1 | 2026-05-03 |
| 61 | Gemma 3 27B | 10 | 2 | 2026-08-23 | |
| 62 | GPT-4o mini | OpenAI | 9.9 | 1 | 2026-05-30 |
| 63 | Seed Oss 36B Instruct | ByteDance | 9.7 | 1 | 2026-06-15 |
| 64 | Qwen3 32B | Alibaba | 8.6 | 1 | 2026-06-15 |
| 65 | Granite 4.1 8B Base | IBM | 7.9 | 2 | 2026-06-15 |
| 66 | Phi 3 Medium 4K Instruct | Microsoft | 7.6 | 1 | 2026-05-30 |
| 67 | OpenAI o1-mini | OpenAI | 7 | 5 | 2026-06-04 |
| 68 | Granite 4.1 3B Base | IBM | 6.8 | 1 | 2026-06-15 |
| 69 | Granite 4.1 30B | IBM | 6.8 | 1 | 2026-06-15 |
| 70 | Gemma 3 12B | 6.3 | 1 | 2026-08-23 | |
| 71 | Qwen 2.5 14B | Alibaba | 5.4 | 1 | 2026-05-30 |
| 72 | Granite 4.1 8B | IBM | 4.8 | 1 | 2026-06-15 |
| 73 | Gemma 3 4B | 4 | 1 | 2026-08-23 | |
| 74 | Phi 4 | Microsoft | 3 | 2 | 2026-08-23 |
| 75 | Gemma 3 1B | 2.2 | 1 | 2026-08-23 | |
| 76 | ERNIE 4.5 | Baidu | 1.8 | 1 | 2026-08-23 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.