模型排行

排名	變化	模型	得分	置信區間
1	—	GPT Image 1.5 (high)閉源OpenAI	1,265	±11
2	—	Nano Banana 2 (Gemini 3.1 Flash Image Preview)閉源Google	1,258	±9
3	—	Riverflow 2.0閉源Sourceful	1,254	±11
4	—	Nano Banana Pro (Gemini 3 Pro Image)閉源Google	1,215	±11
5	—	FLUX.2 [max]閉源Black Forest Labs	1,200	±11
6	—	Seedream 4.0閉源ByteDance Seed	1,185	±7
7	—	FLUX.2 [pro]閉源Black Forest Labs	1,182	±12
8	—	FLUX.2 [flex]閉源Black Forest Labs	1,180	±11
9	—	Imagen 4 Ultra Preview 0606閉源Google	1,174	±11
10	▲ 1	grok-imagine-image閉源xAI	1,172	±8
11	▼ 1	Seedream 4.5閉源ByteDance Seed	1,171	±9
12	—	Imagen 4 Preview 0606閉源Google	1,170	±12
13	—	Imagen 4 Ultra閉源Google	1,170	±11
14	—	Nano Banana (Gemini 2.5 Flash Image)閉源Google	1,163	±7
15	—	Seedream 3.0閉源ByteDance Seed	1,149	±10
16	—	Wan 2.5 Preview閉源Alibaba	1,148	±12
17	—	ImagineArt 1.5 Preview閉源ImagineArt	1,148	±8
18	—	grok-imagine-image-pro閉源xAI	1,138	±8
19	—	Wan 2.6 Image閉源Alibaba	1,133	±16
20	—	Wan2.6 Text to Image閉源Alibaba	1,133	±9
21	—	Recraft V4 Pro閉源Recraft	1,132	±9
22	—	GPT Image 1 (high)閉源OpenAI	1,128	±11
23	—	Vivago 2.1閉源HiDream	1,126	±11
24	—	Kolors 2.1閉源KlingAI	1,124	±11
25	—	Recraft V4閉源Recraft	1,121	±9
26	—	Seedream 5.0 Lite閉源ByteDance	1,119	±10
27	—	FLUX.1 Kontext [max]閉源Black Forest Labs	1,115	±9
28	—	Vidu Q2閉源Vidu	1,113	±9
29	—	Lucid Origin Ultra閉源Leonardo.Ai	1,111	±11
30	—	Eigen Image閉源Eigen AI	1,107	±8
31	—	Lucid Origin Fast閉源Leonardo.Ai	1,106	±11
32	—	Vivago 2.0閉源HiDream	1,103	±12
33	—	Imagen 3 (v002)閉源Google	1,103	±10
34	—	Imagen 4 Standard閉源Google	1,097	±11
35	—	Reve Image (Halfmoon)閉源Reve	1,090	±13
36	—	FLUX1.1 [pro] Ultra閉源Black Forest Labs	1,088	±12
37	—	Imagen 4 Fast Preview 0606閉源Google	1,082	±11
38	—	FLUX.1 Kontext [pro]閉源Black Forest Labs	1,082	±12
39	—	Dreamina 3.1閉源ByteDance	1,080	±11
40	—	Ideogram 3.0閉源Ideogram	1,076	±13
41	—	FLUX1.1 [pro]閉源Black Forest Labs	1,076	±9
42	—	Recraft V3閉源Recraft	1,073	±9
43	—	GPT Image 1 Mini (medium)閉源OpenAI	1,071	±11
44	—	FLUX.1 [pro]閉源Black Forest Labs	1,068	±8
45	▲ 2	P-Image閉源Pruna AI	1,067	±13
46	▼ 1	Midjourney v7 Alpha閉源Midjourney	1,066	±13
47	▼ 1	Qwen Image Plus 2601閉源Alibaba	1,066	±8
48	—	Imagen 4 Fast閉源Google	1,065	±11
49	—	Luma Photon閉源Luma Labs	1,064	±10
50	—	Ideogram v2閉源Ideogram	1,057	±8
51	—	Midjourney v6閉源Midjourney	1,047	±6
52	—	Image-01閉源MiniMax	1,046	±13
53	—	Ideogram v2 Turbo閉源Ideogram	1,043	±8
54	—	Midjourney v6.1閉源Midjourney	1,041	±7
55	—	Z-Image Base閉源Alibaba	1,040	±9
56	—	MAI Image 1閉源Microsoft Azure	1,030	±9
57	—	Firefly Image 5 Preview閉源Adobe	1,027	±11
58	—	Stable Diffusion 3 Large閉源Stability.ai	1,024	±6
59	▲ 1	Phoenix 1.0 Ultra閉源Leonardo.Ai	1,020	±12
60	▼ 1	Ideogram v1閉源Ideogram	1,020	±8
61	—	Ideogram v2a閉源Ideogram	1,013	±13
62	—	Phoenix 0.9 Ultra閉源Leonardo.Ai	1,012	±10
63	—	Ideogram v2a Turbo閉源Ideogram	1,008	±13
64	—	Gemini 2.0 Flash Preview閉源Google	1,004	±14
65	—	Playground v3 (beta)閉源Playground AI	1,004	±8
66	—	Krea 1閉源Krea	1,001	±11
67	—	Firefly Image 4閉源Adobe	999	±11
68	▲ 1	Phoenix 1.0 Fast閉源Leonardo.Ai	984	±12
69	▼ 1	Recraft 20B閉源Recraft	984	±8
70	—	Luma Photon Flash閉源Luma Labs	971	±9
71	—	Runway Gen-4 Image閉源Runway	971	±12
72	—	DALLE 3 HD閉源OpenAI	953	±6
73	—	Gemini 2.0 Flash Experimental閉源Google	947	±20
74	—	DALLE 3閉源OpenAI	938	±6
75	—	Firefly Image 3閉源Adobe	935	±7
76	—	Grok 2閉源xAI	917	±25
77	—	Stable Diffusion 3 Large Turbo閉源Stability.ai	909	±6
78	—	Amazon Titan G1 v2 (Standard)閉源Amazon	909	±7
79	—	Stable Diffusion 1.6閉源Stability.ai	902	±6
80	—	Amazon Titan G1 (Standard)閉源Amazon	891	±7
81	—	DALLE 2閉源OpenAI	719	±7

其餘 31 個模型

智力進步

每家廠商當時最強的模型在 Artificial Analysis 智力指數上的分數，按模型釋出的月份畫。

漲得最多 · Anthropic +57.2目前最高 · Anthropic 64.6最高與最低相差 18.1 分時間範圍 2025年11月 – 2026年11月

64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi