Chatbot Arena (Text - Hard Prompts (English)): leaderboard
Metric: Arena Score. Source: arena.ai. 400 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 (High) | 1536 |
| 2 | Claude Fable 5 (High) | 1533 |
| 3 | Claude Opus 4.6 | 1529 |
| 4 | Claude Opus 4.7 (High) | 1527 |
| 5 | Gemini 3.8 Flash (High) | 1523 |
| 6 | Claude Opus 4.7 | 1521 |
| 7 | GLM-5.3 (Max) | 1521 |
| 8 | Claude Fable 5.1 (Max) | 1518 |
| 9 | Claude Opus 4.8 (High) | 1516 |
| 10 | Kimi K3 (Max) | 1515 |
| 11 | Claude Opus 5 (High) | 1514 |
| 12 | Claude Opus 5 (Max) | 1512 |
| 13 | Qwen 3.8 Max | 1510 |
| 14 | Muse Spark | 1508 |
| 15 | Claude Sonnet 4.6 | 1507 |
Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-text-hard-prompts-english · How It Works · Data refreshed daily, snapshot 2026-09-19.