Reasoning benchmark

# PIQA leaderboard

> PIQA results for 27 AI models, led by GPT-4o mini at 88.7%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/piqa
- Last updated: 2026-10-10
- Title: PIQA Leaderboard (October 2026): Scores by Model | Noometry

As of October 2026, GPT-4o mini has the highest published PIQA score on Noometry at 88.7%, out of 27 models with results.

Last verified October 10, 2026

## About PIQA

Physical commonsense: choose the better way to accomplish a goal.

- **Category:** [Reasoning](https://noometry.com/best/reasoning)
- **Introduced:** 2019
- **Format:** Binary choice
- **Unit:** Percent (random guessing ≈ 50%)
- **Official site:** [yonatanbisk.com](https://yonatanbisk.com/piqa/)

## Top 15 models

Top models on PIQA

1.  GPT-4o mini 88.7%
2.  Phi-3.5-MoE 88.6%
3.  Gemini 1.5 Flash (May 2024) 87.5%
4.  Llama 3.1-405B 85.9%
5.  Falcon-180B 84.9%
6.  DeepSeek-V3 84.7%
7.  DeepSeek-V2 (MoE-236B, May 2024) 83.9%
8.  Gemma 2 9B 83.7%
9.  Mixtral 8x7B 83.6%
10.  Mistral Nemo 83.5%
11.  Falcon-40B 83%
12.  Mistral 7B 83%
13.  Llama 2-70B 82.8%
14.  Qwen2.5 72B Instruct 82.6%
15.  Nemotron-4 15B 82.4%
16.  808284868890

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## All results

PIQA results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [GPT-4o mini](https://noometry.com/models/gpt-4o-mini) | [OpenAI](https://noometry.com/providers/openai) | 88.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 2 | [Phi-3.5-MoE](https://noometry.com/models/phi-3-5-moe) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 88.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 3 | [Gemini 1.5 Flash (May 2024)](https://noometry.com/models/gemini-1-5-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 87.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 4 | [Llama 3.1-405B](https://noometry.com/models/llama-3-1-405b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 85.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 5 | [Falcon-180B](https://noometry.com/models/falcon-180b) |  [![](/logos/tii.svg) Technology Innovation Institute](https://noometry.com/providers/tii) | 84.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 6 | [DeepSeek-V3](https://noometry.com/models/deepseek-v3) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 84.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 7 | [DeepSeek-V2 (MoE-236B, May 2024)](https://noometry.com/models/deepseek-v2) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 83.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 8 | [Gemma 2 9B](https://noometry.com/models/gemma-2-9b) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 83.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 9 | [Mixtral 8x7B](https://noometry.com/models/mixtral-8x7b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 83.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 10 | [Mistral Nemo](https://noometry.com/models/mistral-nemo) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 83.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 11 | [Falcon-40B](https://noometry.com/models/falcon-40b) |  [![](/logos/tii.svg) Technology Innovation Institute](https://noometry.com/providers/tii) | 83% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 12 | [Mistral 7B](https://noometry.com/models/mistral-7b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 83% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 13 | [Llama 2-70B](https://noometry.com/models/llama-2-70b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 82.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 14 | [Qwen2.5 72B Instruct](https://noometry.com/models/qwen2-5-72b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 82.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 15 | [Nemotron-4 15B](https://noometry.com/models/nemotron-4-15b) |  [![](/logos/nvidia.svg) NVIDIA](https://noometry.com/providers/nvidia) | 82.4% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 16 | [Llama 2-34B](https://noometry.com/models/llama-2-34b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 81.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 17 | [Gemma 7B](https://noometry.com/models/gemma-7b) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 81.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 18 | [Llama 3.1-8B](https://noometry.com/models/llama-3-1-8b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 81.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 19 | [Phi-3.5-mini](https://noometry.com/models/phi-3-5-mini) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 81% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 20 | [Llama 2-13B](https://noometry.com/models/llama-2-13b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 80.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 21 | [Falcon-7B](https://noometry.com/models/falcon-7b) |  [![](/logos/tii.svg) Technology Innovation Institute](https://noometry.com/providers/tii) | 80.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 22 | [Llama 13b](https://noometry.com/models/llama-13b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 80.1% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 23 | [Qwen-14B](https://noometry.com/models/qwen-14b) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 79.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 24 | [Llama 2-7B](https://noometry.com/models/llama-2-7b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 78.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 25 | [Qwen-7B](https://noometry.com/models/qwen-7b) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 77.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 26 | [Gemma 2B](https://noometry.com/models/gemma-2b) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 77.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 27 | [Dolly 2.0-12b](https://noometry.com/models/dolly-2-0-12b) |  [![](/logos/databricks.svg) Databricks](https://noometry.com/providers/databricks) | 75.4% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

## Compare the leaders

-   [GPT-4o mini vs Phi-3.5-MoE](https://noometry.com/compare/gpt-4o-mini-vs-phi-3-5-moe)
-   [GPT-4o mini vs Gemini 1.5 Flash (May 2024)](https://noometry.com/compare/gemini-1-5-flash-vs-gpt-4o-mini)
-   [GPT-4o mini vs Llama 3.1-405B](https://noometry.com/compare/gpt-4o-mini-vs-llama-3-1-405b)
-   [GPT-4o mini vs Falcon-180B](https://noometry.com/compare/falcon-180b-vs-gpt-4o-mini)
-   [Phi-3.5-MoE vs Gemini 1.5 Flash (May 2024)](https://noometry.com/compare/gemini-1-5-flash-vs-phi-3-5-moe)
-   [Phi-3.5-MoE vs Llama 3.1-405B](https://noometry.com/compare/llama-3-1-405b-vs-phi-3-5-moe)

## Other reasoning benchmarks

-   [ARC-AGI-2](https://noometry.com/benchmarks/arc-agi-2)
-   [SimpleBench](https://noometry.com/benchmarks/simplebench)
-   [Kagi LLM Benchmark](https://noometry.com/benchmarks/kagi-reasoning)
-   [NYT Connections (extended)](https://noometry.com/benchmarks/nyt-connections)
-   [ARC-AGI-1](https://noometry.com/benchmarks/arc-agi-1)
-   [CritPt](https://noometry.com/benchmarks/critpt)
-   [Chess Puzzles](https://noometry.com/benchmarks/chess-puzzles)
-   [EnigmaEval](https://noometry.com/benchmarks/enigmaeval)
-   [Thematic Generalization](https://noometry.com/benchmarks/thematic-generalization)
-   [LMArena Hard Prompts](https://noometry.com/benchmarks/arena-hard-prompts)
-   [EBR-Bench](https://noometry.com/benchmarks/ebr-bench)
-   [LiveBench Reasoning](https://noometry.com/benchmarks/livebench-reasoning)

## Frequently asked questions

### What does PIQA measure?

Physical commonsense: choose the better way to accomplish a goal.

### Which model has the highest PIQA score?

As of October 2026, GPT-4o mini has the highest published PIQA score on Noometry at 88.7%, out of 27 models with results.

### What is the best open-weight model on PIQA?

Phi-3.5-MoE has the highest PIQA accuracy among open-weight models at 88.6%, ranking 2 of 27 overall.

### Cite this page

Noometry. (2026). PIQA leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/piqa

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/piqa.md).