Reasoning benchmark

# WinoGrande leaderboard

> WinoGrande results for 43 AI models, led by Llama 3.1-405B at 89.2%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/winogrande
- Last updated: 2026-10-10
- Title: WinoGrande Leaderboard (October 2026): Scores by Model

As of October 2026, Llama 3.1-405B has the highest published WinoGrande score on Noometry at 89.2%, out of 43 models with results.

Last verified October 10, 2026

## About WinoGrande

Pronoun-resolution problems that need commonsense reasoning.

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

## Top 15 models

Top models on WinoGrande

1.  Llama 3.1-405B 89.2%
2.  Claude 3 Opus 88.5%
3.  GPT-4 87.5%
4.  GPT-4 87.5%
5.  Falcon-180B 87.1%
6.  DeepSeek-V2 (MoE-236B, May 2024) 86.3%
7.  DeepSeek-V3 85.2%
8.  Deepseek Coder v2 83.7%
9.  Llama 3-70B 83.5%
10.  Qwen2.5 72B Instruct 82.3%
11.  GPT-3.5-turbo 81.6%
12.  phi-3-medium 14B 81.5%
13.  Phi 3 Small 8k Instruct 81.5%
14.  Qwen2.5-Coder-32B 80.8%
15.  Llama 2-70B 80.2%
16.  75808590

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

## All results

WinoGrande results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Llama 3.1-405B](https://noometry.com/models/llama-3-1-405b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 89.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 2 | [Claude 3 Opus](https://noometry.com/models/claude-3-opus) | [Anthropic](https://noometry.com/providers/anthropic) | 88.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 3 | [GPT-4](https://noometry.com/models/gpt-4) | [OpenAI](https://noometry.com/providers/openai) | 87.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 4 | [GPT-4](https://noometry.com/models/gpt-4) | [OpenAI](https://noometry.com/providers/openai) | 87.5% |  | [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) | 87.1% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 6 | [DeepSeek-V2 (MoE-236B, May 2024)](https://noometry.com/models/deepseek-v2) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 86.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 7 | [DeepSeek-V3](https://noometry.com/models/deepseek-v3) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 85.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 8 | [Deepseek Coder v2](https://noometry.com/models/deepseek-coder-v2) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 83.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 9 | [Llama 3-70B](https://noometry.com/models/llama-3-70b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 83.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 10 | [Qwen2.5 72B Instruct](https://noometry.com/models/qwen2-5-72b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 82.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 11 | [GPT-3.5-turbo](https://noometry.com/models/gpt-3-5-turbo) | [OpenAI](https://noometry.com/providers/openai) | 81.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 12 | [phi-3-medium 14B](https://noometry.com/models/phi-3-medium-14b) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 81.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 13 | [Phi 3 Small 8k Instruct](https://noometry.com/models/phi-3-small-8k-instruct) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 81.5% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 14 | [Qwen2.5-Coder-32B](https://noometry.com/models/qwen2-5-coder-32b) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 80.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 15 | [Llama 2-70B](https://noometry.com/models/llama-2-70b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 80.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 16 | [Gemma 7B](https://noometry.com/models/gemma-7b) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 79% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 17 | [Falcon 2 11B](https://noometry.com/models/falcon-2-11b) |  [![](/logos/tii.svg) Technology Innovation Institute](https://noometry.com/providers/tii) | 78.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 18 | [Nemotron-4 15B](https://noometry.com/models/nemotron-4-15b) |  [![](/logos/nvidia.svg) NVIDIA](https://noometry.com/providers/nvidia) | 78% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 19 | [Mixtral 8x7B](https://noometry.com/models/mixtral-8x7b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 77.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 20 | [Falcon-40B](https://noometry.com/models/falcon-40b) |  [![](/logos/tii.svg) Technology Innovation Institute](https://noometry.com/providers/tii) | 76.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 21 | [Llama 2-34B](https://noometry.com/models/llama-2-34b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 76.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 22 | [Llama 3-8B](https://noometry.com/models/llama-3-8b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 75.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 23 | [Mistral 7B](https://noometry.com/models/mistral-7b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 75.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 24 | [Claude 3 Sonnet](https://noometry.com/models/claude-3-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 75.1% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 25 | [Claude 3 Haiku](https://noometry.com/models/claude-3-haiku) | [Anthropic](https://noometry.com/providers/anthropic) | 74.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 26 | [Phi-1.5](https://noometry.com/models/phi-1-5) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 73.4% | 5 | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 27 | [Llama 13b](https://noometry.com/models/llama-13b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 73% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 28 | [Qwen2.5-Coder (1.5B)](https://noometry.com/models/qwen2-5-coder) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 72.9% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 29 | [Llama 2-13B](https://noometry.com/models/llama-2-13b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 72.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 30 | [Yi 6B](https://noometry.com/models/yi-6b) | [01.AI](https://noometry.com/providers/01-ai) | 71.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 31 | [Phi 3 Mini 4k Instruct](https://noometry.com/models/phi-3-mini-4k-instruct) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 70.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 32 | [Llama 2-7B](https://noometry.com/models/llama-2-7b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 69.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 33 | [Falcon-7B](https://noometry.com/models/falcon-7b) |  [![](/logos/tii.svg) Technology Innovation Institute](https://noometry.com/providers/tii) | 67.2% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 34 | [INTELLECT-1](https://noometry.com/models/intellect-1) |  [![](/logos/huggingface.svg) Hugging Face](https://noometry.com/providers/huggingface) | 65.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 35 | [Gemma 2B](https://noometry.com/models/gemma-2b) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 65.4% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 36 | [StarCoder 2 15B](https://noometry.com/models/starcoder-2-15b) |  [![](/logos/nvidia.svg) NVIDIA](https://noometry.com/providers/nvidia) | 64.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 37 | [DeepSeek Coder 33B](https://noometry.com/models/deepseek-coder-33b) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 62% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 38 | [Dolly 2.0-12b](https://noometry.com/models/dolly-2-0-12b) |  [![](/logos/databricks.svg) Databricks](https://noometry.com/providers/databricks) | 61.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 39 | [DeepSeek Coder 6.7B](https://noometry.com/models/deepseek-coder-6-7b) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 57.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 40 | [StarCoder 2 3B](https://noometry.com/models/starcoder-2-3b) |  [![](/logos/nvidia.svg) NVIDIA](https://noometry.com/providers/nvidia) | 57.1% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 41 | [StarCoder 2 7B](https://noometry.com/models/starcoder-2-7b) |  [![](/logos/nvidia.svg) NVIDIA](https://noometry.com/providers/nvidia) | 57.1% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 42 | [Phi-2](https://noometry.com/models/phi-2) |  [![](/logos/microsoft.svg) Microsoft](https://noometry.com/providers/microsoft) | 54.7% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 43 | [DeepSeek Coder 1.3B](https://noometry.com/models/deepseek-coder-1-3b) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 53.3% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

## Compare the leaders

-   [Llama 3.1-405B vs Claude 3 Opus](https://noometry.com/compare/claude-3-opus-vs-llama-3-1-405b)
-   [Llama 3.1-405B vs GPT-4](https://noometry.com/compare/gpt-4-vs-llama-3-1-405b)
-   [Llama 3.1-405B vs GPT-4](https://noometry.com/compare/gpt-4-vs-llama-3-1-405b)
-   [Llama 3.1-405B vs Falcon-180B](https://noometry.com/compare/falcon-180b-vs-llama-3-1-405b)
-   [Claude 3 Opus vs GPT-4](https://noometry.com/compare/claude-3-opus-vs-gpt-4)
-   [Claude 3 Opus vs GPT-4](https://noometry.com/compare/claude-3-opus-vs-gpt-4)

## 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 WinoGrande measure?

Pronoun-resolution problems that need commonsense reasoning.

### Which model has the highest WinoGrande score?

As of October 2026, Llama 3.1-405B has the highest published WinoGrande score on Noometry at 89.2%, out of 43 models with results.

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

Llama 3.1-405B has the highest WinoGrande accuracy among open-weight models at 89.2%, ranking 1 of 43 overall.

### Cite this page

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

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