Chatbot Arena (Text - Coding): leaderboard
The arena.ai text board restricted to battles whose prompt is flagged as code related by a heuristic that looks for code blocks, programming language keywords and commands. Voters compare two anonymous replies to one prompt and pick the better one. The Bradley-Terry rating is refit over this subset alone, and higher is better.
Metric: Arena Score. Source: arena.ai. 409 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 1562 | #8 (Gemini 4 Argon) |
| 2 | Claude Fable 5 (High) | 1551 | #16 (Claude Fable 5) |
| 3 | Claude Opus 4.6 (High) | 1550 | #63 (Claude Opus 4.6) |
| 4 | Claude Opus 4.7 (High) | 1550 | #49 (Claude Opus 4.7) |
| 5 | Claude Opus 5.5 (High) | 1550 | #3 (Claude Opus 5.5) |
| 6 | Claude Opus 4.6 | 1548 | #63 |
| 7 | Claude Opus 4.7 | 1546 | #49 |
| 8 | GPT-6.1 Sol (Max) | 1545 | #7 (GPT-6.1 Sol) |
| 9 | GPT-6 Astra (Max) | 1542 | #6 (GPT-6 Astra) |
| 10 | Kimi K3 (Max) | 1540 | #30 (Kimi K3) |
| 11 | MiMo-V2.6-Pro | 1539 | #42 |
| 12 | Muse Spark 1.3 (Max) | 1538 | #24 (Muse Spark 1.3) |
| 13 | GPT-5.6 Sol (xHigh) | 1535 | #19 (GPT-5.6 Sol) |
| 14 | Claude Opus 5 (High) | 1534 | #12 (Claude Opus 5) |
| 15 | Claude Fable 5.1 (Max) | 1534 | #9 (Claude Fable 5.1) |
Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-text-coding · How It Works · Data refreshed daily, snapshot 2026-10-20.