LMArena Text to Image: 3D modeling
1,246.11 points
Rank 5 · 4,252 samples · Retrieved August 26, 2026
LMArenaImage modelPreview
MAI-Image-2.5
Microsoft AI
- Released
- June 2026
- Data date
- August 15, 2026
Category view
Position within the category
This overview uses only published data from matching cohorts. Missing values never change a rank.
This position gives equal weight to 4 fixed Gradually image tests. One archived first output contributes for each model and task.
- Position
- Rank 3 of 28
- Index score
- 67.6 / 100
- Coverage
- 4 / 4
Leaderboard
1MAI-Image-2.5 Flash76.6 / 100
2GPT Image 271 / 100
3MAI-Image-2.5, Current model67.6 / 100
3Seedream 5.0 Pro67.6 / 100
28Stable Diffusion 3.5 Large2.5 / 100
Measurements
Comparable benchmark results
Each chart contains exactly one source, one measurement series, and one stored comparison cohort. Bars show the position. The measured value appears on the right.
Text to Image Arena
1,303 Elo · Rank 5 of 25
Sample: 14,800 · Data date: August 26, 2026
1.GPT Image 21,371 Elo
3.Nano Banana 21,321 Elo
4.GPT Image 1.51,310 Elo
5.MAI-Image-2.5, Current model1,303 Elo
6.Nano Banana Pro1,298 Elo
7.Nano Banana 2 Lite1,288 Elo
25.FLUX.2 Klein 9B1,144 Elo
7 of 25 model versions shown in this chart. A higher value ranks first.
Source: Artificial AnalysisParticipants: 25
Image Editing Arena
1,257 Elo · Rank 2 of 19
Sample: 11,305 · Data date: August 26, 2026
1.Reve 2.11,263 Elo
2.GPT Image 21,257 Elo
2.MAI-Image-2.5, Current model1,257 Elo
4.GPT Image 1.51,251 Elo
5.Nano Banana 21,250 Elo
19.FLUX.2 Flex1,162 Elo
6 of 19 model versions shown in this chart. A higher value ranks first.
Source: Artificial AnalysisParticipants: 19
Gradually sample test: Typography and layout
93.8 rank points · Rank 1 of 28
Sample: 1 · Data date: August 18, 2026
1.MAI-Image-2.5, Current model93.8 rank points
1.Nano Banana Pro93.8 rank points
3.Recraft V4.1 Utility92.6 rank points
28.Stable Diffusion 3.5 Large0 rank points
4 of 28 model versions shown in this chart. A higher value ranks first.
One archived, unfiltered first output per model. All 28 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-BildtestParticipants: 28
Gradually sample test: Product photography
86.4 rank points · Rank 3 of 28
Sample: 1 · Data date: August 18, 2026
1.GPT Image 295.1 rank points
2.MAI-Image-2.5 Flash93.8 rank points
3.MAI-Image-2.5, Current model86.4 rank points
4.Nano Banana Pro82.7 rank points
5.Krea 2 Large81.5 rank points
28.Stable Diffusion 3.5 Large3.7 rank points
6 of 28 model versions shown in this chart. A higher value ranks first.
One archived, unfiltered first output per model. All 28 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-BildtestParticipants: 28
Gradually sample test: Character and detail
24.7 rank points · Rank 23 of 28
Sample: 1 · Data date: August 18, 2026
1.MAI-Image-2.5 Flash86.4 rank points
21.HiDream-O1-Image27.1 rank points
21.Krea 2 Medium Turbo27.1 rank points
23.MAI-Image-2.5, Current model24.7 rank points
24.Nano Banana Pro21 rank points
25.Nano Banana 218.5 rank points
28.Stable Diffusion 3.5 Large3.7 rank points
7 of 28 model versions shown in this chart. A higher value ranks first.
One archived, unfiltered first output per model. All 28 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-BildtestParticipants: 28
Gradually sample test: Infographic
65.4 rank points · Rank 10 of 28
Sample: 1 · Data date: August 18, 2026
1.Reve 2.197.5 rank points
8.HiDream-O1-Image-1.574.1 rank points
9.Seedream 4.067.9 rank points
10.MAI-Image-2.5, Current model65.4 rank points
11.Luma UNI 1 Max64.2 rank points
12.Nano Banana 2 Lite58 rank points
28.Firefly Image Model 51.2 rank points
7 of 28 model versions shown in this chart. A higher value ranks first.
One archived, unfiltered first output per model. All 28 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-BildtestParticipants: 28
Profile
Specifications and access
Published information about this model. Unknown values are not estimated.
Pricing
Published prices
Prices remain tied to their documented unit and source.
Measurements
Other published benchmarks
The stored dataset does not contain an exactly matching comparison cohort for these values.
LMArena Text to Image: Art
1,277.64 points
Rank 3 · 6,571 samples · Retrieved August 26, 2026
LMArenaLMArena Text to Image: Cartoon
1,270.07 points
Rank 7 · 17,440 samples · Retrieved August 26, 2026
LMArenaLMArena Text to Image: Commercial design
1,260.49 points
Rank 6 · 17,261 samples · Retrieved August 26, 2026
LMArenaLMArena Text to Image: Overall
1,256.13 points
Rank 8 · 44,940 samples · Retrieved August 26, 2026
LMArenaLMArena Text to Image: Photorealism
1,253.01 points
Rank 10 · 17,468 samples · Retrieved August 26, 2026
LMArenaLMArena Text to Image: Portraits
1,259.2 points
Rank 10 · 8,541 samples · Retrieved August 26, 2026
LMArenaLMArena Text to Image: Text rendering
1,282.79 points
Rank 7 · 15,799 samples · Retrieved August 26, 2026
LMArenaLMArena Image Editing: Overall
1,401.28 points
Rank 5 · 170,301 samples · Retrieved August 26, 2026
LMArenaHead-to-head comparisons
Compare this model
Each matchup compares this model with exactly one other model from the same category.
More models
Models from the same selection
All AI models
Evidence
Primary sources and data date
Every statement links to its underlying documentation or leaderboard.