Benchmarks · composite
Arena Text · Industry: life, physical and social science
Crowdsourced pairwise human preference on open-ended text prompts (style control on); Bradley-Terry rating. Only the Industry: life, physical and social science category.
211 models · 222 results · updated 2026-11-09 · arena.ai ↗
Leaderboard
Gemini 4 Argon high
All results
222 rows| # | Model | Effort | Score |
|---|---|---|---|
| 1 | Gemini 4 Argon | high | 1536 ±12 |
| 2 | Opus 4.6 | high | 1527 ±6 |
| 3 | Opus 4.7 | high | 1525 ±7 |
| 4 | Fable 5 | high | 1525 ±8 |
| 5 | Muse Wonder | max | 1524 ±14 |
| 6 | Gemini 4 Carbon | high | 1522 ±24 |
| 7 | Fable 5.1 | max | 1522 ±11 |
| 8 | Opus 5.5 | high | 1521 ±13 |
| 9 | Kimi K3 | max | 1519 ±8 |
| 10 | Opus 4.6 | max | 1518 ±6 |
| 11 | Gemini 3.8 Flash | high | 1518 ±7 |
| 12 | Qwen 3.8 Max | max | 1515 ±9 |
| 13 | Opus 5 | max | 1515 ±9 |
| 14 | Muse Spark 1.3 | max | 1515 ±10 |
| 15 | Opus 4.7 | max | 1514 ±7 |
| 16 | Gemini 3.7 Flash | high | 1513 ±9 |
| 17 | Muse Spark 1.2 | xhigh | 1512 ±13 |
| 18 | GLM 5.3 | max | 1510 ±9 |
| 19 | Opus 5 | high | 1509 ±7 |
| 20 | Gemini 3.1 Pro | high | 1509 ±5 |
| 21 | Opus 4.8 | high | 1504 ±7 |
| 22 | Muse Spark 1.1 | max | 1503 ±8 |
| 23 | GPT-6.1 Sol | max | 1502 ±13 |
| 24 | Sonnet 4.6 | max | 1500 ±7 |
| 25 | Opus 4.8 | max | 1499 ±7 |
| 26 | GPT-5.5 | high | 1498 ±7 |
| 27 | GLM 5.2 | max | 1497 ±7 |
| 28 | Gemini 3 Pro | high | 1497 ±8 |
| 29 | Gemini 3.5 Flash | medium | 1496 ±8 |
| 30 | GPT-5.5 | xhigh | 1496 ±7 |