LMArena Document leaderboard
- Claude Opus 5 leads
- 1516
- models measured
- 29
- lead over #2
- +9 Elo
- median
- 1457
Leader on LMArena Document · read October 21
On October 21, 2026, 29 of our catalog's models had a LMArena Document result, and Anthropic Claude Opus 5 led with 1516, ahead of Anthropic Claude Opus 4.6 at 1507.
Questions about a document the person uploads. People ask a question, compare two anonymous answers and vote for the better one; the rating is built from those votes, as LMArena publishes it.
Role in the index
LMArena Document is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Votes | Interval (±) | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5Anthropic | 1516 | 8,794 | ±7.0 | 83.0 | $25.00 | |
| 2 | Claude Opus 4.6Anthropic | 1507 | 27,929 | ±6.4 | 57.2 | $25.00 | |
| 3 | Claude Fable 5Anthropic | 1496 | 8,497 | ±7.9 | 79.3 | $50.00 | |
| 4 | Claude Opus 4.7Anthropic | 1495 | 22,192 | ±6.4 | 67.5 | $25.00 | |
| 5 | GPT-5.5OpenAI | 1486 | 21,279 | ±6.2 | 71.3 | $30.00 | |
| 6 | GPT-5.6 SolOpenAI | 1483 | 4,818 | ±8.8 | 73.5 | $20.00 | |
| 7 | Claude Sonnet 4.6Anthropic | 1482 | 57,813 | ±5.9 | 57.3 | $15.00 | |
| 8 | Claude Opus 4.8Anthropic | 1475 | 11,551 | ±6.9 | 65.0 | $25.00 | |
| 9 | GPT-5.6 TerraOpenAI | 1472 | 5,787 | ±8.3 | 63.9 | $12.00 | |
| 10 | GPT-5.4OpenAI | 1471 | 33,331 | ±6.0 | 64.4 | $15.00 | |
| 11 | Claude Sonnet 5Anthropic | 1466 | 7,411 | ±7.6 | 57.7 | $10.00 | |
| 12 | Muse Spark 1.1Meta | 1465 | 4,885 | ±8.8 | 70.3 | $4.25 | |
| 13 | Gemini 3.5 FlashGoogle | 1463 | 6,500 | ±7.9 | 39.8 | $9.00 | |
| 14 | Claude Opus 4.5Anthropic | 1462 | 7,981 | ±10.4 | 44.0 | $25.00 | |
| 15 | GPT-5.6 LunaOpenAI | 1457 | 5,766 | ±8.2 | 58.0 | $1.20 | |
| 16 | Gemini 3.6 FlashGoogle | 1456 | 2,975 | ±10.8 | 62.5 | $3.75 | |
| 17 | Grok 4.6xAI | 1452 | 2,258 | ±12.0 | 65.7 | $6.00 | |
| 18 | Grok 4.5xAI | 1452 | 4,965 | ±8.7 | 58.6 | $6.00 | |
| 19 | Kimi K2.6Moonshot AI | 1451 | 11,291 | ±7.6 | 52.1 | $4.00 | |
| 20 | Claude Sonnet 4.5Anthropic | 1450 | 31,455 | ±6.1 | 39.5 | $15.00 | |
| 21 | Qwen3.7 PlusQwen | 1444 | 4,591 | ±9.0 | 33.2 | $1.28 | |
| 22 | Gemini 3.1 Pro PreviewGoogle | 1444 | 49,457 | ±5.2 | 32.3 | $12.00 | |
| 23 | MiniMax M3MiniMax | 1435 | 6,314 | ±8.3 | 39.2 | $1.20 | |
| 24 | Gemma 4 31BGoogle | 1425 | 12,326 | ±7.8 | 39.0 | $0.40 | |
| 25 | Gemini 2.5 ProGoogle | 1421 | 25,110 | ±6.3 | 19.1 | $10.00 | |
| 26 | Claude Haiku 4.5Anthropic | 1420 | 34,677 | ±5.9 | 35.6 | $5.00 | |
| 27 | GLM 5V TurboZ.ai | 1416 | 6,995 | ±8.2 | Not benchmarked | $4.00 | |
| 28 | GPT-5.2OpenAI | 1405 | 7,108 | ±9.5 | 55.9 | $14.00 | |
| 29 | GPT-5.1OpenAI | 1403 | 8,244 | ±9.3 | 40.4 | $10.00 |
Only models in our catalog are listed, and each appears once with its best published run. 184 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.