TriviaQA leaderboard
As of 6 Nov 2026, the top score on TriviaQA is 87.6% by Llama 2-70B (Meta), ahead of Claude 2.0 at 87.5%, across 36 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 18 Jul 2023). Trivia questions with evidence documents. TriviaQA is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- Llama 2-70B holds the top TriviaQA score at 87.6% (Epoch AI Benchmarking Hub, 18 Jul 2023).
- Claude 2.0 is second on TriviaQA at 87.5% (Epoch AI Benchmarking Hub, 11 Jul 2023).
- TriviaQA: Trivia questions with evidence documents. A 2017 benchmark, saturated for current models; kept for history (Epoch AI Benchmarking Hub).
- 36 model configurations have a published TriviaQA result on BenchLeader (BenchLeader, 6 Nov 2026).
- TriviaQA does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Trivia questions with evidence documents. A 2017 benchmark, saturated for current models; kept for history.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Knowledge
- Index weight
- Reference only
- Models
- 36
- Data as of
- 6 Nov 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
What the test looks like
Trivia questions with evidence documents.
How it is scored
Accuracy, as compiled by Epoch AI from published results.
What to keep in mind
Saturated: frontier models score near the ceiling, so it separates only older or smaller models, and results are largely self-reported by developers.
- 1Llama 2-70B87.6%
- 2Claude 2.087.5%
- 3Claude 1.386.7%
- 4PaLM 2-L86.1%
- 5LLaMA-65B86.0%
- 6GPT 3.5 Turbo 110685.8%
- 7GPT 4 061384.8%
- 8Llama 2-34B84.6%
- 9LLaMA-33B83.8%
- 10DeepSeek V382.9%
- 11Llama 3.1 405B82.7%
- 12Mixtral 8x7B82.2%
- 13PaLM 2-M81.7%
- 14PaLM 540B81.4%
- 15Deepseek v280.0%
| # | ||||
|---|---|---|---|---|
| 1 |  Llama 2-70Bopen ↗ | 87.6% | 33.6 | 2023-07-18 |
| 2 |  Claude 2.0 | 87.5% | – | 2023-07-11 |
| 3 | Claude 1.3 | 86.7% | – | 2023-04-18 |
| 4 |  PaLM 2-L | 86.1% | – | 2023-05-17 |
| 5 | LLaMA-65Bopen | 86.0% | – | 2023-02-24 |
| 6 |  GPT 3.5 Turbo 1106 | 85.8% | 38.6 | 2023-11-06 |
| 7 | GPT 4 0613 | 84.8% | 39.5 | 2023-06-13 |
| 8 | Llama 2-34B | 84.6% | – | 2023-07-18 |
| 9 | LLaMA-33B | 83.8% | – | 2023-02-24 |
| 10 |  DeepSeek V3open ↗ | 82.9% | 44.4 | 2024-12-26 |
| 11 | Llama 3.1 405Bopen ↗ | 82.7% | 41.9 | 2024-07-23 |
| 12 |  Mixtral 8x7Bopen | 82.2% | 30.9 | 2023-12-11 |
| 13 | PaLM 2-M | 81.7% | – | 2023-05-17 |
| 14 | PaLM 540B | 81.4% | – | 2022-04-04 |
| 15 | Deepseek v2open ↗ | 80.0% | – | 2024-05-07 |
| 16 |  Falcon-40B | 79.9% | – | 2023-03-15 |
| 17 | Llama 2-13Bopen ↗ | 79.6% | – | 2023-07-18 |
| 18 | Claude Instant 1.1 | 78.9% | – | – |
| 19 | Claude Instant 1.2 | 78.7% | – | 2023-08-09 |
| 20 | LLaMA-13B | 77.9% | – | 2023-02-24 |
| 21 | Mistral 7B v0.3open | 75.2% | 29.0 | 2023-09-27 |
| 22 | PaLM 2-S | 75.2% | – | 2023-05-17 |
| 23 |  Phi 3medium | 73.9% | 37.9 | 2024-04-23 |
| 24 | Llama 2-7Bopen ↗ | 73.7% | – | 2023-07-18 |
| 25 |  MPT-30B | 73.6% | – | 2023-06-22 |
| 26 | Gemma 7B | 72.3% | – | 2024-02-21 |
| 27 |  Qwen2.5 72Bopen ↗ | 71.9% | 43.2 | 2024-09-19 |
| 28 | InstructGPT 175B | 71.2% | – | 2022-01-27 |
| 29 | LLaMA-7B | 71.0% | – | 2023-02-24 |
| 30 | Llama 3-8Bopen | 67.7% | 34.9 | 2024-04-18 |
| 31 | Falcon-7B | 64.6% | – | 2023-04-24 |
| 32 | Phi 3 Miniopen | 64.0% | 36.9 | 2024-04-23 |
| 33 | MPT-7B | 61.6% | – | 2023-05-05 |
| 34 | Phi 3 Small | 58.1% | – | 2024-04-23 |
| 35 | Gemma 2B | 53.2% | – | 2024-02-21 |
| 36 | Phi-2open ↗ | 45.2% | – | 2023-12-12 |
More knowledge benchmarks · All LLM benchmarks
Cite as: BenchLeader, “TriviaQA leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/triviaqa.
TriviaQA: questions
- What does TriviaQA measure?
- Trivia questions with evidence documents. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on TriviaQA?
- As of 6 Nov 2026, the highest score on TriviaQA is 87.6% by Llama 2-70B (result dated 18 Jul 2023), ahead of Claude 2.0 at 87.5%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads TriviaQA?
- Llama 2-70B leads TriviaQA with 87.6% as of 6 Nov 2026 (result dated 18 Jul 2023), ahead of Claude 2.0 at 87.5%.
- How many models have TriviaQA results?
- 36 model configurations have a TriviaQA result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs TriviaQA and how often is it updated?
- TriviaQA is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does TriviaQA count toward the BenchLeader Index?
- No. TriviaQA is shown for reference but left out of the composite index.