Document Arena

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

Sep 13, 2026

404,715 votes

44 models

	Rank Spread
1	14	Claude Opus 5 (High) Anthropic · Proprietary	1516±7	8,794	$5 / $25	1M
2	17	Claude Fable 5.1 (Max) Anthropic · Proprietary	1513±15	1,403	$10 / $50	1M
3	17	Claude Opus 4.6 (High) Anthropic · Proprietary	1507±6	27,929	$5 / $25	1M
4	17	Claude Opus 4.6 Anthropic · Proprietary	1507±6	41,260	$5 / $25	1M
5	211	Claude Fable 5 (High) Anthropic · Proprietary	1496±8	8,497	$10 / $50	1M
6	211	Claude Opus 4.7 Anthropic · Proprietary	1495±6	22,192	$5 / $25	1M
7	211	Claude Opus 4.7 (High) Anthropic · Proprietary	1495±7	21,957	$5 / $25	1M
8	515	GPT-5.5 OpenAI · Proprietary	1486±6	21,279	$5 / $30	1.1M
9	515	GPT-5.5 (High) OpenAI · Proprietary	1484±6	20,944	$5 / $30	1.1M
10	516	GPT-5.6 Sol (xHigh) OpenAI · Proprietary	1483±9	4,818	$4 / $20	N/A
11	816	Claude Sonnet 4.6 Anthropic · Proprietary	1482±6	57,813	$1.50 / $7.50	1M
12	822	Claude Opus 4.8 (High) Anthropic · Proprietary	1475±7	11,551	$5 / $25	1M
13	825	GPT-5.6 Terra (xHigh) OpenAI · Proprietary	1472±8	5,787	$2 / $12	N/A
14	1024	GPT-5.4 OpenAI · Proprietary	1471±6	33,331	$2.50 / $15	1.1M
15	530	Muse Spark 1.3 (Max) Meta · Proprietary	1471±18	1,006	N/A	N/A
16	829	GPT-6 Astra (Max) OpenAI · Proprietary	1468±14	1,621	N/A	N/A
17	1227	Claude Sonnet 5 (High) Anthropic · Proprietary	1466±8	7,411	$2 / $10	1M
18	1229	Muse Spark 1.1 Meta · Proprietary	1465±9	4,885	$1.25 / $4.25	1M
19	1228	Claude Opus 4.8 Anthropic · Proprietary	1464±7	12,128	$5 / $25	1M
20	1229	Gemini 3.5 Flash (Medium) Google · Proprietary	1463±8	6,500	$0.75 / $4.50	1M
21	1230	Claude Opus 4.5 (20251101) Anthropic · Proprietary	1462±10	7,981	$5 / $25	200K
22	1230	Gemini 3.5 Flash (High) Google · Proprietary	1461±9	4,061	$0.75 / $4.50	1M
23	1331	GPT-5.6 Luna (xHigh) OpenAI · Proprietary	1457±8	5,766	$0.20 / $1.20	N/A
24	1331	Gemini 3.6 Flash (High) Google · Proprietary	1456±11	2,975	$0.75 / $3.75	1M
25	1433	Grok 4.6 (High) SpaceXAI · Proprietary	1452±12	2,258	$2 / $6	500K
26	1531	Grok 4.5 SpaceXAI · Proprietary	1452±9	4,965	$2 / $6	500K
27	1631	Kimi K2.6 Moonshot · Modified MIT	1451±8	11,291	$0.95 / $4	262.1K
28	1731	Claude Sonnet 4.5 (20250929) Anthropic · Proprietary	1450±6	31,455	$3 / $15	1M
29	1536	Muse Spark Meta · Proprietary	1444±18	1,084	N/A	N/A
30	2034	Qwen3.7-Plus Alibaba · Proprietary	1444±9	4,591	$0.32 / $1.28	1M
31	2333	Gemini 3.1 Pro Preview Google · Proprietary	1444±5	49,457	$1 / $6	1M
32	2836	MiniMax-M3 MiniMax · MiniMax Community License	1435±8	6,314	$0.30 / $1.20	524.3K
33	2837	Gemini 3 Pro Google · Proprietary	1434±9	10,769	$2 / $12	1M
34	3038	Kimi K2.5 (Thinking) Moonshot · Modified MIT	1430±7	21,569	$0.60 / $3	N/A
35	3140	Gemma 4 31B Google · Apache 2.0	1425±8	12,326	$0.99 / $1.49	131.1K
36	3140	Gemini 2.5 Pro Google · Proprietary	1421±6	25,110	$1.25 / $10	1M
37	3340	Claude Haiku 4.5 (20251001) Anthropic · Proprietary	1420±6	34,677	$1 / $5	200K
38	3543	Grok 4.20 Beta (Reasoning, 0309) SpaceXAI · Proprietary	1416±7	20,948	$1.25 / $2.50	1M
39	3443	GLM-5V-Turbo Z.ai · Proprietary	1416±8	6,995	$1.20 / $4	202.8K
40	3544	Gemini 3 Flash Google · Proprietary	1413±9	7,158	$0.50 / $3	1M
41	3844	GPT-5.2 (High) OpenAI · Proprietary	1405±10	7,108	$0.88 / $7	400K
42	3844	GPT-5.1 OpenAI · Proprietary	1403±9	8,244	$0.63 / $5	400K
43	3844	GPT-5.5 Instant OpenAI · Proprietary	1403±8	8,497	$5 / $30	1.1M
44	4044	GPT-5.2 OpenAI · Proprietary	1401±6	28,212	$0.88 / $7	400K

Document Arena

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

Remove Style Control Leaderboard Plots

Document Arena

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

Remove Style Control Leaderboard Plots

Document Arena

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

Remove Style Control Leaderboard Plots

Fraction of Model A Wins in A vs. B Battles (Ties Excluded)

Average Win Rate Against Each Opponent (Ties Excluded)

Battle Count for Each Pair of Models (Ties Excluded)

Confidence Intervals on Model Strength (via Bootstrapping)

Document Arena

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

Remove Style Control Leaderboard Plots

Fraction of Model A Wins in A vs. B Battles (Ties Excluded)

Average Win Rate Against Each Opponent (Ties Excluded)

Battle Count for Each Pair of Models (Ties Excluded)

Confidence Intervals on Model Strength (via Bootstrapping)