Document Arena

View overall rankings across AI models in document analysis and long-content reasoning.

Jul 26, 2026

322,650 votes

38 models

	Rank Spread
1	17	claude-fable-5 Anthropic · Proprietary	1507±9	5,792	$10 / $50	1M
2	111	claude-opus-4-7 Anthropic · Proprietary	1497±7	18,990	$5 / $25	1M
3	112	claude-opus-4-7-thinking Anthropic · Proprietary	1496±7	18,793	$5 / $25	1M
4	112	claude-opus-4-6 Anthropic · Proprietary	1495±7	37,271	$5 / $25	1M
5	112	claude-opus-4-6-thinking Anthropic · Proprietary	1493±7	24,973	$5 / $25	1M
6	117	claude-opus-5-high Anthropic · Proprietary	1493±15	1,663	$5 / $25	1M
7	217	claude-opus-4-8-thinking Anthropic · Proprietary	1487±8	8,388	$5 / $25	1M
8	221	muse-spark-1.1 Meta · Proprietary	1483±13	2,054	$1.25 / $4.25	1M
9	221	gpt-5.6-terra-xhigh OpenAI · Proprietary	1482±13	1,969	N/A	N/A
10	220	claude-opus-4-8 Anthropic · Proprietary	1482±8	8,193	$5 / $25	1M
11	123	gpt-5.6-sol-xhigh OpenAI · Proprietary	1481±17	1,152	N/A	N/A
12	321	claude-sonnet-5-high Anthropic · Proprietary	1479±11	3,601	$2 / $10	1M
13	621	gpt-5.5-high OpenAI · Proprietary	1478±7	16,816	$5 / $30	1.1M
14	621	claude-sonnet-4-6 Anthropic · Proprietary	1478±6	54,427	$3 / $15	1M
15	621	gpt-5.5 OpenAI · Proprietary	1475±7	17,286	$5 / $30	1.1M
16	823	gpt-5.4 OpenAI · Proprietary	1469±7	29,809	$2.50 / $15	1.1M
17	627	gpt-5.6-luna-xhigh OpenAI · Proprietary	1467±14	1,888	N/A	N/A
18	629	muse-spark Meta · Proprietary	1466±18	1,078	N/A	N/A
19	826	claude-opus-4-5-20251101 Anthropic · Proprietary	1464±10	7,965	$5 / $25	200K
20	928	gemini-3.5-flash-medium Google · Proprietary	1462±10	3,689	$1.50 / $9	1M
21	829	grok-4.5 SpaceXAI · Proprietary	1462±12	2,435	$2 / $6	500K
22	1526	gemini-3.1-pro-preview Google · Proprietary	1460±6	45,162	$2 / $12	1M
23	1530	gpt-5.5-instant OpenAI · Proprietary	1455±9	8,442	$5 / $30	1.1M
24	1732	gemini-3-pro Google · Proprietary	1451±9	10,739	$2 / $12	1M
25	1732	kimi-k2.6 Moonshot · Modified MIT	1450±8	11,181	$0.95 / $4	262.1K
26	1732	qwen3.7-plus Alibaba · Proprietary	1449±11	3,088	$0.32 / $1.28	1M
27	2032	claude-sonnet-4-5-20250929 Anthropic · Proprietary	1446±7	28,634	$3 / $15	200K
28	1932	gemma-4-31b Google · Apache 2.0	1445±8	10,577	N/A	N/A
29	2135	gemini-3-flash Google · Proprietary	1441±9	7,173	$0.50 / $3	1M
30	2334	grok-4.20-beta-0309-reasoning SpaceXAI · Proprietary	1440±8	18,666	$2 / $6	2M
31	2436	kimi-k2.5-thinking Moonshot · Modified MIT	1435±7	19,891	$0.60 / $3	N/A
32	2436	minimax-m3 MiniMax · MiniMax Community License	1434±8	6,263	$0.60 / $2.40	N/A
33	2936	gemini-2.5-pro Google · Proprietary	1430±6	24,963	$1.25 / $10	1M
34	3037	gpt-5.2 OpenAI · Proprietary	1426±6	28,068	$1.75 / $14	400K
35	2937	gpt-5.2-high OpenAI · Proprietary	1424±10	7,073	$1.75 / $14	400K
36	3137	claude-haiku-4-5-20251001 Anthropic · Proprietary	1422±7	30,937	$1 / $5	200K
37	3438	gpt-5.1 OpenAI · Proprietary	1412±9	8,220	$1.25 / $10	400K
38	3738	glm-5v-turbo Z.ai · Proprietary	1402±10	4,894	$1.20 / $4	202.8K

Default Leaderboard Plots

Confidence Intervals on Model Strength (via Bootstrapping)

Fraction of Model A Wins for All Non-tied A vs. B Battles

Battle Count for Each Combination of Models (without Ties)

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)

Rank Spread

claude-fable-5

Anthropic · Proprietary

1507±9

5,792

$10 / $50

111

claude-opus-4-7

Anthropic · Proprietary

1497±7

18,990

$5 / $25

112

claude-opus-4-7-thinking

Anthropic · Proprietary

1496±7

18,793

$5 / $25

112

claude-opus-4-6

Anthropic · Proprietary

1495±7

37,271

$5 / $25

112

claude-opus-4-6-thinking

Anthropic · Proprietary

1493±7

24,973

$5 / $25

117

claude-opus-5-high

Anthropic · Proprietary

1493±15

1,663

$5 / $25

217

claude-opus-4-8-thinking

Anthropic · Proprietary

1487±8

8,388

$5 / $25

221

muse-spark-1.1

Meta · Proprietary

1483±13

2,054

$1.25 / $4.25

221

gpt-5.6-terra-xhigh

OpenAI · Proprietary

1482±13

1,969

N/A

220

claude-opus-4-8

Anthropic · Proprietary

1482±8

8,193

$5 / $25

123

gpt-5.6-sol-xhigh

OpenAI · Proprietary

1481±17

1,152

N/A

321

claude-sonnet-5-high

Anthropic · Proprietary

1479±11

3,601

$2 / $10

621

gpt-5.5-high

OpenAI · Proprietary

1478±7

16,816

$5 / $30

1.1M

621

claude-sonnet-4-6

Anthropic · Proprietary

1478±6

54,427

$3 / $15

621

gpt-5.5

OpenAI · Proprietary

1475±7

17,286

$5 / $30

1.1M

823

gpt-5.4

OpenAI · Proprietary

1469±7

29,809

$2.50 / $15

1.1M

627

gpt-5.6-luna-xhigh

OpenAI · Proprietary

1467±14

1,888

N/A

629

muse-spark

Meta · Proprietary

1466±18

1,078

N/A

826

claude-opus-4-5-20251101

Anthropic · Proprietary

1464±10

7,965

$5 / $25

200K

928

gemini-3.5-flash-medium

Google · Proprietary

1462±10

3,689

$1.50 / $9

829

grok-4.5

SpaceXAI · Proprietary

1462±12

2,435

$2 / $6

500K

1526

gemini-3.1-pro-preview

Google · Proprietary

1460±6

45,162

$2 / $12

1530

gpt-5.5-instant

OpenAI · Proprietary

1455±9

8,442

$5 / $30

1.1M

1732

gemini-3-pro

Google · Proprietary

1451±9

10,739

$2 / $12

1732

kimi-k2.6

Moonshot · Modified MIT

1450±8

11,181

$0.95 / $4

262.1K

1732

qwen3.7-plus

Alibaba · Proprietary

1449±11

3,088

$0.32 / $1.28

2032

claude-sonnet-4-5-20250929

Anthropic · Proprietary

1446±7

28,634

$3 / $15

200K

1932

gemma-4-31b

Google · Apache 2.0

1445±8

10,577

N/A

2135

gemini-3-flash

Google · Proprietary

1441±9

7,173

$0.50 / $3

2334

grok-4.20-beta-0309-reasoning

SpaceXAI · Proprietary

1440±8

18,666

$2 / $6

2436

kimi-k2.5-thinking

Moonshot · Modified MIT

1435±7

19,891

$0.60 / $3

N/A

2436

minimax-m3

MiniMax · MiniMax Community License

1434±8

6,263

$0.60 / $2.40

N/A

2936

gemini-2.5-pro

Google · Proprietary

1430±6

24,963

$1.25 / $10

3037

gpt-5.2

OpenAI · Proprietary

1426±6

28,068

$1.75 / $14

400K

2937

gpt-5.2-high

OpenAI · Proprietary

1424±10

7,073

$1.75 / $14

400K

3137

claude-haiku-4-5-20251001

Anthropic · Proprietary

1422±7

30,937

$1 / $5

200K

3438

gpt-5.1

OpenAI · Proprietary

1412±9

8,220

$1.25 / $10

400K

3738

glm-5v-turbo

Z.ai · Proprietary

1402±10

4,894

$1.20 / $4

202.8K