Updated daily at 05:00 UTC
Benchmark
Document Arena (arena.ai)
Measures: Users upload documents and ask two anonymous models to analyse or reason over the long content, then vote; votes are turned into an Elo-scale score with a confidence interval (Style Control off by default).
- Maintainer
- Arena Intelligence (arena.ai; formerly LMArena / LMSYS Chatbot Arena)
- Introduced
- —
- Status
- Active
- Saturation
- Headroom
- Typical provenance
- Independent
- In scoring set
- Yes
Leaderboard
Independent and third-party results only.
| Model | Arena score (Elo scale) | Score CI (+/-) | Votes | Provenance | Source | Evaluated |
|---|---|---|---|---|---|---|
| Claude Fable 5 | 1499 | — | 8497 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.6 Anthropic · High effort | 1495 | — | 27929 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.7 | 1494 | — | 22192 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.6 | 1494 | — | 41260 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.7 Anthropic · High effort | 1493 | — | 21957 | Independent | Arena | 13 Sep 2026 |
| Claude Fable 5.1 Anthropic · Max effort | 1493 | — | 1403 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 5 Anthropic · High effort | 1490 | — | 8794 | Independent | Arena | 13 Sep 2026 |
| GPT-5.6 Sol OpenAI · Xhigh effort | 1489 | — | 4818 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.8 Anthropic · High effort | 1486 | — | 11551 | Independent | Arena | 13 Sep 2026 |
| GPT-5.5 | 1483 | — | 21279 | Independent | Arena | 13 Sep 2026 |
| GPT-5.5 OpenAI · High effort | 1480 | — | 20944 | Independent | Arena | 13 Sep 2026 |
| Muse Spark 1.1 | 1479 | — | 4885 | Independent | Arena | 13 Sep 2026 |
| Claude Sonnet 4.6 | 1478 | — | 57813 | Independent | Arena | 13 Sep 2026 |
| GPT-5.6 Terra OpenAI · Xhigh effort | 1477 | — | 5787 | Independent | Arena | 13 Sep 2026 |
| Claude Sonnet 5 Anthropic · High effort | 1476 | — | 7411 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.8 | 1475 | — | 12128 | Independent | Arena | 13 Sep 2026 |
| GPT-6 Astra OpenAI · Max effort | 1473 | — | 1621 | Independent | Arena | 13 Sep 2026 |
| GPT-5.4 | 1470 | — | 33331 | Independent | Arena | 13 Sep 2026 |
| Muse Spark 1.3 Meta · Max effort | 1468 | — | 1006 | Independent | Arena | 13 Sep 2026 |
| Claude Opus 4.5 | 1465 | — | 7981 | Independent | Arena | 13 Sep 2026 |
| Grok 4.5 | 1464 | — | 4965 | Independent | Arena | 13 Sep 2026 |
| GPT-5.6 Luna OpenAI · Xhigh effort | 1462 | — | 5766 | Independent | Arena | 13 Sep 2026 |
| Grok 4.6 SpaceXAI (xAI) · High effort | 1461 | — | 2258 | Independent | Arena | 13 Sep 2026 |
| Gemini 3.5 Flash Google DeepMind · Medium effort | 1461 | — | 6500 | Independent | Arena | 13 Sep 2026 |
| Gemini 3.1 Pro | 1459 | — | 49457 | Independent | Arena | 13 Sep 2026 |
| Gemini 3.6 Flash Google DeepMind · High effort | 1453 | — | 2975 | Independent | Arena | 13 Sep 2026 |
| Gemini 3.5 Flash Google DeepMind · High effort | 1453 | — | 4061 | Independent | Arena | 13 Sep 2026 |
| Qwen3.7-Plus | 1453 | — | 4591 | Independent | Arena | 13 Sep 2026 |
| Gemini 3 Pro | 1451 | — | 10769 | Independent | Arena | 13 Sep 2026 |
| Kimi K2.6 | 1450 | — | 11291 | Independent | Arena | 13 Sep 2026 |
| Claude Sonnet 4.5 | 1449 | — | 31455 | Independent | Arena | 13 Sep 2026 |
| Gemma 4 31B | 1445 | — | 12326 | Independent | Arena | 13 Sep 2026 |
| Grok 4.20 SpaceXAI (xAI) · Reasoning effort | 1439 | — | 20948 | Independent | Arena | 13 Sep 2026 |
| Gemini 3 Flash | 1439 | — | 7158 | Independent | Arena | 13 Sep 2026 |
| Kimi K2.5 Moonshot AI · Thinking effort | 1436 | — | 21569 | Independent | Arena | 13 Sep 2026 |
| MiniMax-M3 | 1433 | — | 6314 | Independent | Arena | 13 Sep 2026 |
| Gemini 2.5 Pro | 1431 | — | 25110 | Independent | Arena | 13 Sep 2026 |
| GPT-5.2 | 1425 | — | 28212 | Independent | Arena | 13 Sep 2026 |
| GPT-5.2 OpenAI · High effort | 1423 | — | 7108 | Independent | Arena | 13 Sep 2026 |
| Claude Haiku 4.5 | 1419 | — | 34677 | Independent | Arena | 13 Sep 2026 |
| GPT-5.1 | 1416 | — | 8244 | Independent | Arena | 13 Sep 2026 |
| GLM-5V-Turbo | 1400 | — | 6995 | Independent | Arena | 13 Sep 2026 |
- Arena (arena.ai), lmarena-ai/leaderboard-dataset, CC BY 4.0.