# Arena Elo (Vision) — AI model rankings

Real people give two anonymous AIs the same image — a photo, a screenshot, a diagram — and vote for whichever reads it better. The votes become a chess-style Elo rating on arena.ai. It measures which AI people find more useful at looking at things, not how it scores on a fixed test set. Higher is better.

31 tracked models have a published Arena Elo (Vision) score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Claude Fable 5 | Anthropic | 1312 | Jun 9 2026 |
| 2 | Qwen3.8-Max | Qwen | 1302 | Aug 3 2026 |
| 3 | Claude Opus 4.7 | Anthropic | 1301 | Apr 16 2026 |
| 4 | Claude Opus 4.6 | Anthropic | 1299 | Feb 5 2026 |
| 5 | Muse Spark | Meta | 1294 | Apr 8 2026 |
| 6 | Claude Opus 5 | Anthropic | 1292 | Jul 24 2026 |
| 6 | Muse Spark 1.2 | Meta | 1292 | Aug 5 2026 |
| 8 | Gemini 3.0 Pro | Google | 1289 | Nov 18 2025 |
| 9 | Gemini 3.5 Flash | Google | 1286 | May 19 2026 |
| 10 | GPT-5.5 | OpenAI | 1285 | Apr 23 2026 |
| 10 | Claude Opus 4.8 | Anthropic | 1285 | May 28 2026 |
| 10 | Gemini 3.6 Flash | Google | 1285 | Jul 21 2026 |
| 13 | Grok 4.5 | SpaceXAI | 1282 | Jul 8 2026 |
| 14 | GPT-5.6 Sol | OpenAI | 1281 | Jun 26 2026 |
| 14 | Muse Spark 1.1 | Meta | 1281 | Jul 9 2026 |
| 16 | GPT-5.4 | OpenAI | 1280 | Mar 5 2026 |
| 17 | GPT-5.2 | OpenAI | 1278 | Dec 11 2025 |
| 18 | Gemini 3.1 Pro | Google | 1277 | Feb 19 2026 |
| 19 | Claude Sonnet 4.6 | Anthropic | 1275 | Feb 17 2026 |
| 20 | Gemini 3.0 Flash | Google | 1271 | Dec 17 2025 |
| 20 | Claude Sonnet 5 | Anthropic | 1271 | Jun 30 2026 |
| 22 | Gemini 3.5 Flash-Lite | Google | 1270 | Jul 21 2026 |
| 23 | GPT-5.6 Terra | OpenAI | 1266 | Jun 26 2026 |
| 24 | Qwen3.7-Plus | Qwen | 1265 | Jun 1 2026 |
| 25 | Kimi K2.6 | Moonshot AI | 1263 | Apr 21 2026 |
| 26 | GPT-5.6 Luna | OpenAI | 1253 | Jun 26 2026 |
| 26 | Qwen3.8-27B | Qwen | 1253 | Aug 14 2026 |
| 28 | GPT-5.4 mini | OpenAI | 1252 | Mar 17 2026 |
| 29 | GPT-5.1 | OpenAI | 1250 | Nov 12 2025 |
| 29 | Kimi K2.5 | Moonshot AI | 1250 | Jan 27 2026 |
| 31 | Gemini 2.5 Pro | Google | 1246 | Mar 25 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/arena-vision-elo
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Benchmark scores are the figures published by the releasing lab at launch.