TriviaQA — Generic/unlabeled TriviaQA rollup; MiMo/Moonlight cards
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Hunyuan Large | Tencent | 89.2 | 1 | 2026-05-31 |
| 2 | Claude 2 | Anthropic | 87.5 | 1 | 2026-07-17 |
| 3 | Claude 1.3 | Anthropic | 86.7 | 1 | 2026-07-17 |
| 4 | DeepSeek-V4-Pro-Base | DeepSeek | 85.6 | 3 | 2026-06-05 |
| 5 | Kimi K2 Base | Moonshot | 85.1 | 6 | 2026-08-23 |
| 6 | DeepSeek-V3.2-Exp-Base | DeepSeek | 83.9 | 2 | 2026-06-13 |
| 7 | Gemma 2 27B IT | 83.7 | 1 | 2026-08-23 | |
| 8 | DeepSeek-V3.1-Base | DeepSeek | 83.5 | 2 | 2026-06-13 |
| 9 | DeepSeek-V4-Flash-Base | DeepSeek | 82.8 | 3 | 2026-06-05 |
| 10 | MiMo V2.5 Pro | Xiaomi | 81.3 | 1 | 2026-08-23 |
| 11 | MiMo V2.5 Pro Base | Xiaomi | 81.3 | 3 | 2026-06-05 |
| 12 | MiMo V2.5 Base | Xiaomi | 80.7 | 3 | 2026-06-05 |
| 13 | MiMo V2 Flash Base | Xiaomi | 80.3 | 2 | 2026-06-13 |
| 14 | DeepSeek-V2 | DeepSeek | 79.9 | 1 | 2026-05-31 |
| 15 | Gemma 2 Instruct (9B) | 76.6 | 1 | 2026-08-23 | |
| 16 | Mistral Large 3 | Mistral | 74.9 | 1 | 2026-08-23 |
| 17 | Moonlight | Moonshot | 66.3 | 2 | 2026-06-12 |
| 18 | Llama 3.2 3B Instruct | Meta | 59.6 | 2 | 2026-06-05 |
| 19 | Qwen2.5 3B | Alibaba | 51.1 | 2 | 2026-06-05 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.