OpenBookQA leaderboard
As of 6 Nov 2026, the top score on OpenBookQA is 88.0%, shared by Phi 3 Mini and Phi 3 Small, ahead of Phi 3 at 87.4%, across 41 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 23 Apr 2024). Elementary science questions with an open book of facts. OpenBookQA is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- Phi 3 Mini and Phi 3 Small share the top OpenBookQA score at 88.0% (Epoch AI Benchmarking Hub, 6 Nov 2026).
- OpenBookQA: Elementary science questions with an open book of facts. A 2018 benchmark, saturated for current models; kept for history (Epoch AI Benchmarking Hub).
- 41 model configurations have a published OpenBookQA result on BenchLeader (BenchLeader, 6 Nov 2026).
- OpenBookQA does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Elementary science questions with an open book of facts. A 2018 benchmark, saturated for current models; kept for history.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Knowledge
- Index weight
- Reference only
- Models
- 41
- 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
Elementary science questions with an open book of facts.
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.
- 1Phi 3 Mini88.0%
- 2Phi 3 Small88.0%
- 3Phi 3 (medium)87.4%
- 4GPT 3.5 Turbo 110686.0%
- 5Mixtral 8x7B85.8%
- 6Llama 3-8B82.6%
- 7Mistral 7B v0.379.8%
- 8Gemma 7B78.6%
- 9Phi-273.6%
- 10PaLM 540B68.0%
- 11InstructGPT 175B65.4%
- 12Falcon-180B64.2%
- 13Llama 2-70B60.2%
- 14LLaMA-65B60.2%
- 15Llama 2-7B58.6%
| # | ||||
|---|---|---|---|---|
| 1 |  Phi 3 Miniopen | 88.0% | 36.9 | 2024-04-23 |
| 2 | Phi 3 Small | 88.0% | – | 2024-04-23 |
| 3 | Phi 3medium | 87.4% | 37.9 | 2024-04-23 |
| 4 |  GPT 3.5 Turbo 1106 | 86.0% | 38.6 | 2023-11-06 |
| 5 |  Mixtral 8x7Bopen | 85.8% | 30.9 | 2023-12-11 |
| 6 |  Llama 3-8Bopen | 82.6% | 34.9 | 2024-04-18 |
| 7 | Mistral 7B v0.3open | 79.8% | 29.0 | 2023-09-27 |
| 8 |  Gemma 7B | 78.6% | – | 2024-02-21 |
| 9 | Phi-2open ↗ | 73.6% | – | 2023-12-12 |
| 10 | PaLM 540B | 68.0% | – | 2022-04-04 |
| 11 | InstructGPT 175B | 65.4% | – | 2022-01-27 |
| 12 |  Falcon-180B | 64.2% | – | 2023-09-06 |
| 13 | Llama 2-70Bopen ↗ | 60.2% | 33.6 | 2023-07-18 |
| 14 | LLaMA-65Bopen | 60.2% | – | 2023-02-24 |
| 15 | Llama 2-7Bopen ↗ | 58.6% | – | 2023-07-18 |
| 16 | LLaMA-33B | 58.6% | – | 2023-02-24 |
| 17 | Llama 2-34B | 58.2% | – | 2023-07-18 |
| 18 | PaLM 2-M | 57.4% | – | 2023-05-17 |
| 19 | LLaMA-7B | 57.2% | – | 2023-02-24 |
| 20 | Llama 2-13Bopen ↗ | 57.0% | – | 2023-07-18 |
| 21 | Falcon-40B | 56.6% | – | 2023-03-15 |
| 22 | LLaMA-13B | 56.4% | – | 2023-02-24 |
| 23 | PaLM 2-S | 56.2% | – | 2023-05-17 |
| 24 |  MPT-30B | 52.0% | – | 2023-06-22 |
| 25 | Falcon-7B | 51.6% | – | 2023-04-24 |
| 26 | MPT-7B | 51.4% | – | 2023-05-05 |
| 27 | PaLM 62B | 50.4% | – | 2022-04-04 |
| 28 |  Xgen 7B Base | 40.2% | – | 2023-06-27 |
| 29 |  RedPajama-INCITE-7B-Base | 40.0% | – | 2023-05-04 |
| 30 | OPT-13B | 39.8% | – | 2022-05-11 |
| 31 |  Dolly v2 12B | 39.2% | – | 2023-04-11 |
| 32 | open_llama_7b | 39.0% | – | 2023-06-07 |
| 33 |  GPT-NeoX-20B | 38.8% | – | 2022-04-07 |
| 34 | GPT-J-6B | 38.2% | – | 2021-08-05 |
| 35 | Phi-1.5 | 37.2% | – | 2023-09-11 |
| 36 |  Cerebras-GPT-13B | 35.8% | – | 2023-03-20 |
| 37 |  Vicuna 13B v1.1 | 33.0% | – | 2023-04-12 |
| 38 |  StableLM Tuned Alpha 7B | 32.4% | – | 2023-04-19 |
| 39 | OPT-1.3B | 24.0% | – | 2022-05-11 |
| 40 | GPT-Neo-2.7B | 23.2% | – | 2021-03-30 |
| 41 | Gpt2 Xl | 22.4% | – | 2019-11-05 |
More knowledge benchmarks · All LLM benchmarks
Cite as: BenchLeader, “OpenBookQA leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/openbookqa.
OpenBookQA: questions
- What does OpenBookQA measure?
- Elementary science questions with an open book of facts. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on OpenBookQA?
- As of 6 Nov 2026, the highest score on OpenBookQA is 88.0%, shared by Phi 3 Mini and Phi 3 Small, ahead of Phi 3 at 87.4%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads OpenBookQA?
- Phi 3 Mini and Phi 3 Small tie for first on OpenBookQA at 88.0% as of 6 Nov 2026, ahead of Phi 3 at 87.4%.
- How many models have OpenBookQA results?
- 41 model configurations have a OpenBookQA result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs OpenBookQA and how often is it updated?
- OpenBookQA 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 OpenBookQA count toward the BenchLeader Index?
- No. OpenBookQA is shown for reference but left out of the composite index.