---
title: "Documents: model rankings and source evidence · Undominated.ai"
canonical: https://undominated.ai/rankings/arena-document/
description: "Documents ranked by Arena score. Inspect exact model variants, source dates, measurement limits and downloadable evidence."
---

# Documents: model rankings and source evidence · Undominated.ai

> Documents ranked by Arena score. Inspect exact model variants, source dates, measurement limits and downloadable evidence.

[All rankings](/rankings/) / Search and documents

# Documents

Human preference ratings for the overall category of this Arena task. Scores from different boards are not interchangeable.

**44 entries** · Arena score · Higher is better

Source publication: 2026-09-13. Retrieved 2026-11-05.

Source: [Arena (lmarena-ai)](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset) · [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/).

Ranks use confidence bounds. Several models can share a rank; the rows within a shared rank are displayed by score.

Documents · Arena score

| Significance rank | Model / evaluated variant | Arena score | 95% confidence interval | Votes |
| --- | --- | --- | --- | --- |
| 1 | claude-opus-5-high anthropic | 1,516.27 | 1,509.25 – 1,523.29 | 8,794 |
| 1 | claude-fable-5.1-max anthropic | 1,512.58 | 1,497.4 – 1,527.76 | 1,403 |
| 1 | claude-opus-4-6-high anthropic | 1,507.32 | 1,500.9 – 1,513.74 | 27,929 |
| 1 | claude-opus-4-6 anthropic | 1,507.05 | 1,501.18 – 1,512.92 | 41,260 |
| 2 | claude-fable-5 anthropic | 1,496.08 | 1,488.22 – 1,503.94 | 8,497 |
| 2 | claude-opus-4-7 anthropic | 1,495.04 | 1,488.66 – 1,501.41 | 22,192 |
| 2 | claude-opus-4-7-high anthropic | 1,494.71 | 1,488.21 – 1,501.22 | 21,957 |
| 5 | gpt-5.5 openai | 1,485.72 | 1,479.48 – 1,491.96 | 21,279 |
| 5 | gpt-5.5-high openai | 1,484.48 | 1,478.1 – 1,490.86 | 20,944 |
| 5 | gpt-5.6-sol-xhigh openai | 1,482.71 | 1,473.91 – 1,491.51 | 4,818 |
| 8 | claude-sonnet-4-6 anthropic | 1,482.24 | 1,476.36 – 1,488.11 | 57,813 |
| 8 | claude-opus-4-8-high anthropic | 1,474.79 | 1,467.89 – 1,481.69 | 11,551 |
| 8 | gpt-5.6-terra-xhigh openai | 1,471.58 | 1,463.31 – 1,479.85 | 5,787 |
| 10 | gpt-5.4 openai | 1,471 | 1,464.97 – 1,477.04 | 33,331 |
| 5 | muse-spark-1.3-max meta | 1,470.66 | 1,452.37 – 1,488.95 | 1,006 |
| 8 | gpt-6-astra-max openai | 1,467.52 | 1,453.26 – 1,481.79 | 1,621 |
| 12 | claude-sonnet-5-high anthropic | 1,466.14 | 1,458.49 – 1,473.78 | 7,411 |
| 12 | muse-spark-1.1 meta | 1,464.7 | 1,455.86 – 1,473.54 | 4,885 |
| 12 | claude-opus-4-8 anthropic | 1,464.07 | 1,457.1 – 1,471.03 | 12,128 |
| 12 | gemini-3.5-flash-medium google | 1,463.27 | 1,455.4 – 1,471.14 | 6,500 |
| 12 | claude-opus-4-5-20251101 anthropic | 1,462.2 | 1,451.79 – 1,472.61 | 7,981 |
| 12 | gemini-3.5-flash-high google | 1,461.4 | 1,452.14 – 1,470.66 | 4,061 |
| 13 | gpt-5.6-luna-xhigh openai | 1,456.92 | 1,448.72 – 1,465.11 | 5,766 |
| 13 | gemini-3.6-flash-high google | 1,456.06 | 1,445.28 – 1,466.83 | 2,975 |
| 14 | grok-4.6-high xai | 1,452.33 | 1,440.31 – 1,464.35 | 2,258 |
| 15 | grok-4.5 xai | 1,451.96 | 1,443.26 – 1,460.66 | 4,965 |
| 16 | kimi-k2.6 moonshot | 1,450.74 | 1,443.1 – 1,458.37 | 11,291 |
| 17 | claude-sonnet-4-5-20250929 anthropic | 1,449.81 | 1,443.69 – 1,455.93 | 31,455 |
| 15 | muse-spark meta | 1,444.19 | 1,425.9 – 1,462.48 | 1,084 |
| 20 | qwen3.7-plus alibaba | 1,444.14 | 1,435.12 – 1,453.17 | 4,591 |
| 23 | gemini-3.1-pro-preview google | 1,443.92 | 1,438.69 – 1,449.14 | 49,457 |
| 28 | minimax-m3 minimax | 1,434.79 | 1,426.53 – 1,443.06 | 6,314 |
| 28 | gemini-3-pro google | 1,434.09 | 1,425.18 – 1,443 | 10,769 |
| 30 | kimi-k2.5-thinking moonshot | 1,430.1 | 1,423.37 – 1,436.82 | 21,569 |
| 31 | gemma-4-31b google | 1,424.51 | 1,416.71 – 1,432.31 | 12,326 |
| 31 | gemini-2.5-pro google | 1,421.23 | 1,414.98 – 1,427.49 | 25,110 |
| 33 | claude-haiku-4-5-20251001 anthropic | 1,419.97 | 1,414.1 – 1,425.84 | 34,677 |
| 35 | grok-4.20-beta-0309-reasoning xai | 1,416.18 | 1,409.21 – 1,423.15 | 20,948 |
| 34 | glm-5v-turbo zai | 1,415.59 | 1,407.38 – 1,423.81 | 6,995 |
| 35 | gemini-3-flash google | 1,412.99 | 1,403.59 – 1,422.4 | 7,158 |
| 38 | gpt-5.2-high openai | 1,404.59 | 1,395.09 – 1,414.09 | 7,108 |
| 38 | gpt-5.1 openai | 1,403.17 | 1,393.83 – 1,412.51 | 8,244 |
| 38 | gpt-5.5-instant openai | 1,402.95 | 1,395.02 – 1,410.88 | 8,497 |
| 40 | gpt-5.2 openai | 1,401.12 | 1,395.2 – 1,407.04 | 28,212 |

## What this ranking can tell you

 - The population is the models in this source board, including variants and models without an available API price.
- Ranks use the published confidence bounds: one plus the number of models whose lower bound exceeds this model’s upper bound. Shared ranks are not proof of equal performance.

## Check the source

The original evidence is retained byte for byte. The downloadable board includes the source URLs, revision and SHA-256 hashes.

[Download this board and provenance →](/data/rankings/arena-document.json)

 Source files and licences - [metadata · a7a159f557a1](/data/rankings/evidence/a7a159f557a1e594a5d41218fe8d53c63c63daa39b655e51d6fc5b482a17d059.json) · [upstream source ↗](https://huggingface.co/api/datasets/lmarena-ai/leaderboard-dataset)
- [licence · 88a46d472f35](/data/rankings/evidence/88a46d472f35274fb6b7a852c3d16c8ee8007d51b37f290161537bec26895ee4.txt) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/raw/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/README.md)
- [rows · 5fddd9f56aa8](/data/rankings/evidence/5fddd9f56aa810f58b49a4463d2d400129c814bd7d76e908120956f86ae5540d.parquet) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/resolve/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/document/latest-00000-of-00001.parquet)

[Explore other model types →](/rankings/)[Read the benchmark guide →](/benchmarks/)

## Continue your investigation

 - [Understand the measurements](/benchmarks/)
- [Compare task performance](/best/)
- [Compare embedding models](/embeddings/)
- [Inspect ranking rules](/methodology/)