Chatbot Arena (Text): leaderboard
A user sends a text prompt, two anonymous models answer, and the user votes for the better reply. Model names appear only after the vote, and votes cast once identities are visible do not count. The number is a Bradley-Terry rating fit over all counted text battles, taken from the site's default view, which controls for answer length and markdown formatting, on an Elo-like scale where higher is better.
Metric: Elo. Source: arena.ai. Saturation forecast: Around November 2027. 414 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 1525 | #8 (Gemini 4 Argon) |
| 2 | Claude Opus 5.5 (High) | 1507 | #3 (Claude Opus 5.5) |
| 3 | Claude Opus 4.6 (High) | 1504 | #63 (Claude Opus 4.6) |
| 4 | Claude Fable 5 (High) | 1504 | #16 (Claude Fable 5) |
| 5 | Claude Opus 4.7 (High) | 1501 | #49 (Claude Opus 4.7) |
| 6 | Claude Fable 5.1 (Max) | 1501 | #9 (Claude Fable 5.1) |
| 7 | Claude Opus 4.6 | 1498 | #63 |
| 8 | Gemini 3.8 Flash (High) | 1497 | #21 (Gemini 3.8 Flash) |
| 9 | Claude Opus 4.7 | 1494 | #49 |
| 10 | Muse Spark 1.3 (Max) | 1494 | #24 (Muse Spark 1.3) |
| 11 | Muse Spark 1.2 (xHigh) | 1492 | #62 (Muse Spark 1.2) |
| 12 | Muse Spark 1.1 | 1491 | #52 |
| 13 | Claude Opus 5 (High) | 1490 | #12 (Claude Opus 5) |
| 14 | Claude Opus 5 (Max) | 1489 | #12 (Claude Opus 5) |
| 15 | Muse Spark | 1489 | #64 |
No result here: #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-text · How It Works · Data refreshed daily, snapshot 2026-10-20.