# Arena Elo (Text) — AI model rankings

Real people chat with two anonymous AIs side by side and vote for the answer they prefer. Votes become a chess-style Elo rating on arena.ai — it measures which AI people actually like, not test scores. Higher is better.

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

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Claude Fable 5 | Anthropic | 1509 | Jun 9 2026 |
| 2 | Claude Opus 4.6 | Anthropic | 1504 | Feb 5 2026 |
| 3 | Claude Opus 4.7 | Anthropic | 1503 | Apr 16 2026 |
| 4 | Muse Spark 1.2 | Meta | 1498 | Aug 5 2026 |
| 5 | Claude Opus 5 | Anthropic | 1495 | Jul 24 2026 |
| 6 | Qwen3.8-Max | Qwen | 1491 | Aug 3 2026 |
| 7 | Muse Spark 1.1 | Meta | 1490 | Jul 9 2026 |
| 7 | Gemini 3.7 Flash | Google | 1490 | Aug 13 2026 |
| 9 | Muse Spark | Meta | 1488 | Apr 8 2026 |
| 10 | Gemini 3.0 Pro | Google | 1486 | Nov 18 2025 |
| 10 | Kimi K3 | Moonshot AI | 1486 | Jul 16 2026 |
| 12 | Gemini 3.1 Pro | Google | 1485 | Feb 19 2026 |
| 13 | Claude Opus 4.8 | Anthropic | 1482 | May 28 2026 |
| 13 | Gemini 3.6 Flash | Google | 1482 | Jul 21 2026 |
| 15 | GPT-5.5 | OpenAI | 1481 | Apr 23 2026 |
| 15 | GPT-5.6 Sol | OpenAI | 1481 | Jun 26 2026 |
| 17 | GPT-5.2 | OpenAI | 1476 | Dec 11 2025 |
| 17 | GPT-5.4 | OpenAI | 1476 | Mar 5 2026 |
| 17 | Gemini 3.5 Flash | Google | 1476 | May 19 2026 |
| 20 | Grok 4.20 Beta | SpaceXAI | 1475 | Feb 17 2026 |
| 20 | Qwen3.7-Max | Qwen | 1475 | May 20 2026 |
| 22 | Gemini 3.0 Flash | Google | 1473 | Dec 17 2025 |
| 23 | Claude Sonnet 4.6 | Anthropic | 1472 | Feb 17 2026 |
| 24 | GLM-5.1 | Z.ai | 1468 | Apr 7 2026 |
| 24 | Grok 4.5 | SpaceXAI | 1468 | Jul 8 2026 |
| 26 | GPT-5.6 Terra | OpenAI | 1467 | Jun 26 2026 |
| 27 | Grok 4.1 | SpaceXAI | 1466 | Nov 17 2025 |
| 28 | Claude Sonnet 5 | Anthropic | 1463 | Jun 30 2026 |
| 29 | Kimi K2.6 | Moonshot AI | 1461 | Apr 21 2026 |
| 30 | Gemini 3.5 Flash-Lite | Google | 1459 | Jul 21 2026 |
| 31 | Qwen3.7-Plus | Qwen | 1458 | Jun 1 2026 |
| 32 | DeepSeek-V4-Pro | DeepSeek | 1457 | Apr 24 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/arena-text-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.