Document Arena

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

Jun 5, 2026

235,837 votes

28 models

Rank by

	Rank Spread
1	14	claude-opus-4-6 Anthropic · Proprietary	1509±7	30,103	$5 / $25	1M
2	14	claude-opus-4-6-thinking Anthropic · Proprietary	1507±8	18,865	$5 / $25	1M
3	17	claude-opus-4-7 Anthropic · Proprietary	1499±7	12,863	$5 / $25	1M
4	18	claude-opus-4-7-thinking Anthropic · Proprietary	1495±7	12,625	$5 / $25	1M
5	39	gpt-5.5-high OpenAI · Proprietary	1488±8	10,579	$5 / $30	1.1M
6	39	gpt-5.5 OpenAI · Proprietary	1487±7	10,815	$5 / $30	1.1M
7	39	claude-sonnet-4-6 Anthropic · Proprietary	1487±6	46,925	$3 / $15	1M
8	411	claude-opus-4-8 Anthropic · Proprietary	1476±13	2,214	$5 / $25	1M
9	811	gpt-5.4 OpenAI · Proprietary	1472±7	22,628	$2.50 / $15	1.1M
10	511	claude-opus-4-8-thinking Anthropic · Proprietary	1471±12	2,428	$5 / $25	1M
11	814	claude-opus-4-5-20251101 Anthropic · Proprietary	1461±10	7,994	$5 / $25	200K
12	1117	kimi-k2.6 Moonshot · Modified MIT	1450±9	7,745	$0.95 / $4	262.1K
13	1116	claude-sonnet-4-5-20250929 Anthropic · Proprietary	1450±7	22,899	$3 / $15	200K
14	1122	muse-spark Meta · Proprietary	1440±18	1,044	N/A	N/A
15	1218	gemini-3.1-pro-preview Google · Proprietary	1439±6	35,883	$2 / $12	1M
16	1221	minimax-m3 MiniMax · Proprietary	1438±12	2,684	$0.60 / $2.40	N/A
17	1321	gemini-3-pro Google · Proprietary	1433±9	10,753	$2 / $12	1M
18	1422	kimi-k2.5-thinking Moonshot · Modified MIT	1429±7	15,506	$0.60 / $3	N/A
19	1524	gemma-4-31b Google · Apache 2.0	1424±9	7,308	N/A	N/A
20	1524	gemini-2.5-pro Google · Proprietary	1421±6	24,480	$1.25 / $10	1M
21	1725	claude-haiku-4-5-20251001 Anthropic · Proprietary	1417±7	24,878	$1 / $5	200K
22	1928	gemini-3-flash Google · Proprietary	1413±9	7,196	$0.50 / $3	1M
23	1528	glm-5v-turbo Z.ai · Proprietary	1409±23	646	$1.20 / $4	202.8K
24	1928	grok-4.20-beta-0309-reasoning xAI · Proprietary	1408±8	12,836	$2 / $6	2M
25	2128	gpt-5.2-high OpenAI · Proprietary	1405±9	7,097	$1.75 / $14	400K
26	2228	gpt-5.5-instant OpenAI · Proprietary	1402±8	8,400	$5 / $30	1.1M
27	2228	gpt-5.1 OpenAI · Proprietary	1401±9	8,259	$1.25 / $10	400K
28	2228	gpt-5.2 OpenAI · Proprietary	1401±6	27,604	$1.75 / $14	400K

Document Arena

Remove Style Control Leaderboard Plots

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

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)

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