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	19	Claude Fable 5 (High) Anthropic · Proprietary	1499±8	8,497	$10 / $50	1M
2	110	Claude Opus 4.6 (High) Anthropic · Proprietary	1495±7	27,929	$5 / $25	1M
3	111	Claude Opus 4.7 Anthropic · Proprietary	1494±6	22,192	$5 / $25	1M
4	111	Claude Opus 4.6 Anthropic · Proprietary	1494±6	41,260	$5 / $25	1M
5	113	Claude Opus 4.7 (High) Anthropic · Proprietary	1493±7	21,957	$5 / $25	1M
6	119	Claude Fable 5.1 (Max) Anthropic · Proprietary	1493±15	1,403	$10 / $50	1M
7	118	Claude Opus 5 (High) Anthropic · Proprietary	1490±7	8,794	$5 / $25	1M
8	119	GPT-5.6 Sol (xHigh) OpenAI · Proprietary	1489±9	4,818	$4 / $20	N/A
9	119	Claude Opus 4.8 (High) Anthropic · Proprietary	1486±7	11,551	$5 / $25	1M
10	219	GPT-5.5 OpenAI · Proprietary	1483±6	21,279	$5 / $30	1.1M
11	521	GPT-5.5 (High) OpenAI · Proprietary	1480±7	20,944	$5 / $30	1.1M
12	323	Muse Spark 1.1 Meta · Proprietary	1479±9	4,885	$1.25 / $4.25	1M
13	623	Claude Sonnet 4.6 Anthropic · Proprietary	1478±6	57,813	$1.50 / $7.50	1M
14	626	GPT-5.6 Terra (xHigh) OpenAI · Proprietary	1477±8	5,787	$2 / $12	N/A
15	626	Claude Sonnet 5 (High) Anthropic · Proprietary	1476±8	7,411	$2 / $10	1M
16	726	Claude Opus 4.8 Anthropic · Proprietary	1475±7	12,128	$5 / $25	1M
17	531	GPT-6 Astra (Max) OpenAI · Proprietary	1473±14	1,621	N/A	N/A
18	1128	GPT-5.4 OpenAI · Proprietary	1470±6	33,331	$2.50 / $15	1.1M
19	634	Muse Spark 1.3 (Max) Meta · Proprietary	1468±18	1,006	N/A	N/A
20	634	Muse Spark Meta · Proprietary	1467±18	1,084	N/A	N/A
21	1133	Claude Opus 4.5 (20251101) Anthropic · Proprietary	1465±10	7,981	$5 / $25	200K
22	1232	Grok 4.5 SpaceXAI · Proprietary	1464±9	4,965	$2 / $6	500K
23	1433	GPT-5.6 Luna (xHigh) OpenAI · Proprietary	1462±8	5,766	$0.20 / $1.20	N/A
24	1234	Grok 4.6 (High) SpaceXAI · Proprietary	1461±12	2,258	$2 / $6	500K
25	1433	Gemini 3.5 Flash (Medium) Google · Proprietary	1461±8	6,500	$0.75 / $4.50	1M
26	1434	GPT-5.5 Instant OpenAI · Proprietary	1460±9	8,497	$5 / $30	1.1M
27	1733	Gemini 3.1 Pro Preview Google · Proprietary	1459±5	49,457	$1 / $6	1M
28	1737	Gemini 3.6 Flash (High) Google · Proprietary	1453±11	2,975	$0.75 / $3.75	1M
29	1836	Gemini 3.5 Flash (High) Google · Proprietary	1453±9	4,061	$0.75 / $4.50	1M
30	1836	Qwen3.7-Plus Alibaba · Proprietary	1453±9	4,591	$0.32 / $1.28	1M
31	1837	Gemini 3 Pro Google · Proprietary	1451±9	10,769	$2 / $12	1M
32	1937	Kimi K2.6 Moonshot · Modified MIT	1450±8	11,291	$0.95 / $4	262.1K
33	2036	Claude Sonnet 4.5 (20250929) Anthropic · Proprietary	1449±6	31,455	$3 / $15	1M
34	2439	Gemma 4 31B Google · Apache 2.0	1445±8	12,326	$0.16 / $0.47	128K
35	2840	Grok 4.20 Beta (Reasoning, 0309) SpaceXAI · Proprietary	1439±7	20,948	$1.25 / $2.50	1M
36	2841	Gemini 3 Flash Google · Proprietary	1439±9	7,158	$0.68 / $4.05	1M
37	3141	Kimi K2.5 (Thinking) Moonshot · Modified MIT	1436±7	21,569	$0.60 / $3	N/A
38	3443	MiniMax-M3 MiniMax · MiniMax Community License	1433±8	6,314	$0.30 / $1.20	524.3K
39	3443	Gemini 2.5 Pro Google · Proprietary	1431±6	25,110	$1.25 / $10	1M
40	3643	GPT-5.2 OpenAI · Proprietary	1425±6	28,212	$0.88 / $7	400K
41	3543	GPT-5.2 (High) OpenAI · Proprietary	1423±9	7,108	$0.88 / $7	400K
42	3843	Claude Haiku 4.5 (20251001) Anthropic · Proprietary	1419±6	34,677	$1 / $5	200K
43	3844	GPT-5.1 OpenAI · Proprietary	1416±9	8,244	$0.63 / $5	400K
44	4344	GLM-5V-Turbo Z.ai · Proprietary	1400±8	6,995	$1.20 / $4	202.8K

Document Arena

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

With Style Control Leaderboard Plots

Document Arena

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

With Style Control Leaderboard Plots

Document Arena

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

With 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.

With 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)