Writing & Preference benchmark

# WildBench leaderboard

> WildBench results for 57 AI models, led by Qwen3 235B-A22B at 86.6%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/wildbench
- Last updated: 2026-10-10
- Title: WildBench Leaderboard (October 2026): Scores by Model

As of October 2026, Qwen3 235B-A22B has the highest published WildBench score on Noometry at 86.6%, out of 57 models with results.

Last verified October 10, 2026

## About WildBench

Challenging requests taken from real chatbot conversations, graded by an LLM judge with task-specific checklists. HELM Capabilities run.

- **Category:** [Writing & Preference](https://noometry.com/best/writing)
- **Introduced:** 2024
- **Size:** 1,024 tasks
- **Format:** LLM-judged answers
- **Unit:** Percent (random guessing ≈ 0%)
- **Official site:** [crfm.stanford.edu](https://crfm.stanford.edu/helm/capabilities/latest/)

## Top 15 models

Top models on WildBench

1.  Qwen3 235B-A22B 86.6%
2.  GPT-5.1 86.3%
3.  Kimi K2 (Jul 2025) 86.2%
4.  o3 86.1%
5.  Gemini 3 Pro 85.9%
6.  GPT-5 85.7%
7.  Gemini 2.5 Pro 85.7%
8.  GPT-5 Mini 85.5%
9.  Claude Sonnet 4.5 85.4%
10.  GPT-4.1 85.4%
11.  o4-mini 85.4%
12.  Claude Opus 4 85.2%
13.  Grok 3 84.9%
14.  gpt-oss-120b 84.5%
15.  Claude Haiku 4.5 83.9%
16.  8384858687

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

## All results

WildBench results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Qwen3 235B-A22B](https://noometry.com/models/qwen3-235b-a22b) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 86.6% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 2 | [GPT-5.1](https://noometry.com/models/gpt-5-1) | [OpenAI](https://noometry.com/providers/openai) | 86.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 3 | [Kimi K2 (Jul 2025)](https://noometry.com/models/kimi-k2) | [Moonshot AI](https://noometry.com/providers/moonshot) | 86.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 4 | [o3](https://noometry.com/models/o3) | [OpenAI](https://noometry.com/providers/openai) | 86.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 5 | [Gemini 3 Pro](https://noometry.com/models/gemini-3-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 85.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 6 | [GPT-5](https://noometry.com/models/gpt-5) | [OpenAI](https://noometry.com/providers/openai) | 85.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 7 | [Gemini 2.5 Pro](https://noometry.com/models/gemini-2-5-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 85.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 8 | [GPT-5 Mini](https://noometry.com/models/gpt-5-mini) | [OpenAI](https://noometry.com/providers/openai) | 85.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 9 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 85.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 10 | [GPT-4.1](https://noometry.com/models/gpt-4-1) | [OpenAI](https://noometry.com/providers/openai) | 85.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 11 | [o4-mini](https://noometry.com/models/o4-mini) | [OpenAI](https://noometry.com/providers/openai) | 85.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 12 | [Claude Opus 4](https://noometry.com/models/claude-opus-4) | [Anthropic](https://noometry.com/providers/anthropic) | 85.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 13 | [Grok 3](https://noometry.com/models/grok-3) | [xAI](https://noometry.com/providers/xai) | 84.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 14 | [gpt-oss-120b](https://noometry.com/models/gpt-oss-120b) | [OpenAI](https://noometry.com/providers/openai) | 84.5% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 15 | [Claude Haiku 4.5](https://noometry.com/models/claude-haiku-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 83.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 16 | [Claude Sonnet 4](https://noometry.com/models/claude-sonnet-4) | [Anthropic](https://noometry.com/providers/anthropic) | 83.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 17 | [GPT-4.1 mini](https://noometry.com/models/gpt-4-1-mini) | [OpenAI](https://noometry.com/providers/openai) | 83.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 18 | [DeepSeek-V3](https://noometry.com/models/deepseek-v3) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 83% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 19 | [GPT-4o](https://noometry.com/models/gpt-4o) | [OpenAI](https://noometry.com/providers/openai) | 82.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 20 | [DeepSeek-R1](https://noometry.com/models/deepseek-r1) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 82.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 21 | [Gemini 2.5 Flash-Lite](https://noometry.com/models/gemini-2-5-flash-lite) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 81.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 22 | [Gemini 2.5 Flash](https://noometry.com/models/gemini-2-5-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 81.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 23 | [Claude 3.7 Sonnet](https://noometry.com/models/claude-3-7-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 81.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 24 | [Gemini 1.5 Pro (May 2024)](https://noometry.com/models/gemini-1-5-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 81.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 25 | [GPT-4.1 nano](https://noometry.com/models/gpt-4-1-nano) | [OpenAI](https://noometry.com/providers/openai) | 81.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 26 | [Qwen3-Next 80B-A3B Instruct](https://noometry.com/models/qwen3-next-80b-a3b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 80.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 27 | [GPT-5 Nano](https://noometry.com/models/gpt-5-nano) | [OpenAI](https://noometry.com/providers/openai) | 80.6% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 28 | [Qwen2.5 72B Instruct](https://noometry.com/models/qwen2-5-72b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 80.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 29 | [Mistral Large](https://noometry.com/models/mistral-large) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 80.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 30 | [Gemini 2.0 Flash (Feb 2025)](https://noometry.com/models/gemini-2-0-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 80% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 31 | [Llama 4 Maverick](https://noometry.com/models/llama-4-maverick) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 80% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 32 | [Grok 4](https://noometry.com/models/grok-4) | [xAI](https://noometry.com/providers/xai) | 79.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 33 | [Claude 3.5 Sonnet](https://noometry.com/models/claude-3-5-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 79.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 34 | [Gemini 1.5 Flash (May 2024)](https://noometry.com/models/gemini-1-5-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 79.2% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 35 | [GPT-4o mini](https://noometry.com/models/gpt-4o-mini) | [OpenAI](https://noometry.com/providers/openai) | 79.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 36 | [Gemini 2.0 Flash-Lite](https://noometry.com/models/gemini-2-0-flash-lite) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 79% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 37 | [GLM-4.5-Air](https://noometry.com/models/glm-4-5-air) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 78.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 38 | [Mistral Small 3.1](https://noometry.com/models/mistral-small-3-1) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 78.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 39 | [Nova Premier 1.0](https://noometry.com/models/nova-premier-1-0) | [Amazon](https://noometry.com/providers/amazon) | 78.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 40 | [Llama 3.1-405B](https://noometry.com/models/llama-3-1-405b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 78.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 41 | [Llama 4 Scout](https://noometry.com/models/llama-4-scout) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 78% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 42 | [Amazon Nova Pro](https://noometry.com/models/amazon-nova-pro) | [Amazon](https://noometry.com/providers/amazon) | 77.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 43 | [Claude 3.5 Haiku](https://noometry.com/models/claude-3-5-haiku) | [Anthropic](https://noometry.com/providers/anthropic) | 76% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 44 | [Llama 3.1-70B](https://noometry.com/models/llama-3-1-70b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 75.8% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 45 | [Amazon Nova Lite](https://noometry.com/models/amazon-nova-lite) | [Amazon](https://noometry.com/providers/amazon) | 75% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 46 | [Amazon Nova Micro](https://noometry.com/models/amazon-nova-micro) | [Amazon](https://noometry.com/providers/amazon) | 74.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 47 | [Granite 4.0 H Small](https://noometry.com/models/ibm-granite-h-small) | [IBM](https://noometry.com/providers/ibm) | 73.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 48 | [gpt-oss-20b](https://noometry.com/models/gpt-oss-20b) | [OpenAI](https://noometry.com/providers/openai) | 73.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 49 | [Olmo 2 0325 32b Instruct](https://noometry.com/models/olmo-2-0325-32b-instruct) |  [![](/logos/ai2.svg) Allen Institute for AI (Ai2)](https://noometry.com/providers/ai2) | 73.4% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 50 | [Qwen2.5 7B Instruct](https://noometry.com/models/qwen2-5-7b-instruct) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 73.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 51 | [Mixtral 8x22B](https://noometry.com/models/mixtral-8x22b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 71.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 52 | [OLMo 2 Furious 13B](https://noometry.com/models/olmo-2-furious-13b) |  [![](/logos/ai2.svg) Allen Institute for AI (Ai2)](https://noometry.com/providers/ai2) | 68.9% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 53 | [Llama 3.1-8B](https://noometry.com/models/llama-3-1-8b) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 68.7% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 54 | [Mixtral 8x7B](https://noometry.com/models/mixtral-8x7b) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 67.3% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 55 | [Granite 4.0 Micro](https://noometry.com/models/granite-4-0-micro) | [IBM](https://noometry.com/providers/ibm) | 67% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 56 | [Mistral](https://noometry.com/models/mistral) |  [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 66% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| 57 | [Grok-3 mini](https://noometry.com/models/grok-3-mini) | [xAI](https://noometry.com/providers/xai) | 65.1% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |

## Compare the leaders

-   [Qwen3 235B-A22B vs GPT-5.1](https://noometry.com/compare/gpt-5-1-vs-qwen3-235b-a22b)
-   [Qwen3 235B-A22B vs Kimi K2 (Jul 2025)](https://noometry.com/compare/kimi-k2-vs-qwen3-235b-a22b)
-   [Qwen3 235B-A22B vs o3](https://noometry.com/compare/o3-vs-qwen3-235b-a22b)
-   [Qwen3 235B-A22B vs Gemini 3 Pro](https://noometry.com/compare/gemini-3-pro-vs-qwen3-235b-a22b)
-   [GPT-5.1 vs Kimi K2 (Jul 2025)](https://noometry.com/compare/gpt-5-1-vs-kimi-k2)
-   [GPT-5.1 vs o3](https://noometry.com/compare/gpt-5-1-vs-o3)

## Other writing & preference benchmarks

-   [LMArena Text](https://noometry.com/benchmarks/arena-text)
-   [LMArena Creative Writing](https://noometry.com/benchmarks/arena-creative-writing)
-   [Short-Story Creative Writing](https://noometry.com/benchmarks/lech-mazur-writing)
-   [EQ-Bench Creative Writing](https://noometry.com/benchmarks/eqbench-creative-writing)
-   [LMArena Multi-Turn](https://noometry.com/benchmarks/arena-multi-turn)
-   [EQ-Bench 4](https://noometry.com/benchmarks/eqbench-4)
-   [LiveBench Language](https://noometry.com/benchmarks/livebench-language)

## Frequently asked questions

### What does WildBench measure?

Challenging requests taken from real chatbot conversations, graded by an LLM judge with task-specific checklists. HELM Capabilities run.

### Which model has the highest WildBench score?

As of October 2026, Qwen3 235B-A22B has the highest published WildBench score on Noometry at 86.6%, out of 57 models with results.

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

Qwen3 235B-A22B has the highest WildBench accuracy among open-weight models at 86.6%, ranking 1 of 57 overall.

### Cite this page

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

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