Noometry

Multimodal benchmark

LMArena Document leaderboard

As of October 2026, Claude Opus 5 has the highest published LMArena Document score on Noometry at 1516, out of 38 models with results.

Last verified

About LMArena Document

Blind votes on answers to questions about uploaded documents.

Category
Multimodal
Introduced
2026
Format
Pairwise human votes
Unit
Arena rating (Bradley–Terry)
Official site
lmarena.ai

Top 15 models

  1. Claude Opus 5 1516
  2. Claude Fable 5.1 1513
  3. Claude Opus 4.6 1507
  4. Claude Fable 5 1496
  5. Claude Opus 4.7 1495
  6. GPT-5.5 1486
  7. GPT-5.6 Sol 1483
  8. Claude Sonnet 4.6 1482
  9. Claude Opus 4.8 1475
  10. GPT-5.6 Terra 1472
  11. GPT-5.4 1471
  12. Muse Spark 1.3 1471
  13. GPT-6 Astra 1468
  14. Claude Sonnet 5 1466
  15. Muse Spark 1.1 1465

Sponsored placements are available on pages like this one. Advertise on Noometry

All results

#ModelProviderRatingSettingSourceDate
1Claude Opus 5 Anthropic1516highLMArena2026-09-13
2Claude Fable 5.1 Anthropic1513LMArena2026-09-13
3Claude Opus 4.6 Anthropic1507highLMArena2026-09-13
4Claude Fable 5 Anthropic1496LMArena2026-09-13
5Claude Opus 4.7 Anthropic1495LMArena2026-09-13
6GPT-5.5 OpenAI1486LMArena2026-09-13
7GPT-5.6 Sol OpenAI1483xhighLMArena2026-09-13
8Claude Sonnet 4.6 Anthropic1482LMArena2026-09-13
9Claude Opus 4.8 Anthropic1475highLMArena2026-09-13
10GPT-5.6 Terra OpenAI1472xhighLMArena2026-09-13
11GPT-5.4 OpenAI1471LMArena2026-09-13
12Muse Spark 1.3 Meta1471LMArena2026-09-13
13GPT-6 Astra OpenAI1468LMArena2026-09-13
14Claude Sonnet 5 Anthropic1466highLMArena2026-09-13
15Muse Spark 1.1 Meta1465LMArena2026-09-13
16Gemini 3.5 Flash Google1463mediumLMArena2026-09-13
17Claude Opus 4.5 Anthropic1462LMArena2026-09-13
18GPT-5.6 Luna OpenAI1457xhighLMArena2026-09-13
19Gemini 3.6 Flash Google1456highLMArena2026-09-13
20Grok 4.6 xAI1452highLMArena2026-09-13
21Grok 4.5 xAI1452LMArena2026-09-13
22Kimi K2.6 Moonshot AI1451LMArena2026-09-13
23Claude Sonnet 4.5 Anthropic1450LMArena2026-09-13
24Muse Spark Meta1444LMArena2026-09-13
25Qwen3.7 Plus Alibaba (Qwen)1444LMArena2026-09-13
26Gemini 3.1 Pro Preview Google1444LMArena2026-09-13
27MiniMax-M3 MiniMax1435LMArena2026-09-13
28Gemini 3 Pro Google1434LMArena2026-09-13
29Kimi K2.5 Moonshot AI1430thinkingLMArena2026-09-13
30Gemma 4 31B IT Google1425LMArena2026-09-13
31Gemini 2.5 Pro Google1421LMArena2026-09-13
32Claude Haiku 4.5 Anthropic1420LMArena2026-09-13
33Grok 4.20 (Non-Reasoning) xAI1416LMArena2026-09-13
34GLM-5V-Turbo Z.ai (Zhipu)1416LMArena2026-09-13
35Gemini 3 Flash Preview Google1413LMArena2026-09-13
36GPT-5.2 OpenAI1405highLMArena2026-09-13
37GPT-5.1 OpenAI1403LMArena2026-09-13
38GPT-5.5 Instant OpenAI1403LMArena2026-09-13

Compare the leaders

Other multimodal benchmarks

Frequently asked questions

What does LMArena Document measure?

Blind votes on answers to questions about uploaded documents.

Which model has the highest LMArena Document score?

As of October 2026, Claude Opus 5 has the highest published LMArena Document score on Noometry at 1516, out of 38 models with results.

What is the best open-weight model on LMArena Document?

Kimi K2.6 has the highest LMArena Document rating among open-weight models at 1451, ranking 22 of 38 overall.