TriviaQA: leaderboard
Large-scale reading comprehension dataset with 95K question-answer pairs authored by trivia enthusiasts. Tests factual knowledge retrieval and reasoning over evidence documents.
Metric: Accuracy (%). Source: huggingface.co. Source: epoch.ai. Saturation forecast: Estimated already saturated. 39 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 2 70B Base | 87.6 | #1025 |
| 2 | Claude 2 | 87.5 | #761 |
| 3 | LLaMA-65B | 86 | #1303 |
| 4 | GPT-3.5 Turbo (1106) | 85.8 | #961 |
| 5 | GPT-4 (0613) | 84.8 | #491 |
| 6 | DeepSeek V3 | 82.9 | #321 |
| 7 | Llama 3.1 405B | 82.7 | #466 |
| 8 | Mixtral 8x7B (v0.1) | 82.2 | #1083 |
| 9 | DeepSeek V2 | 80 | #549 |
| 10 | falcon-40B | 79.9 | #1442 |
| 11 | Llama 2 13B | 79.6 | #1436 |
| 12 | Claude Instant 1.2 | 78.7 | #1017 |
| 13 | LLaMA-13B | 77.9 | #1516 |
| 14 | Mistral-7B-v0.1 | 75.2 | #1326 |
| 15 | Phi-3-medium-128k-instruct | 73.9 | #1013 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=triviaqa · How It Works · Data refreshed daily, snapshot 2026-10-20.