MAI-Image-2.6: Benchmarks, Pricing & Comparison

Image samples from our tests

These images come from documented single requests with an archived prompt. They complement measurements but do not replace a controlled laboratory test.

Open Character and detail at original size

Character and detail

This sample tests Anatomy, clothing, and spatial consistency.

Prompt and settings

Create a square cinematic full-body portrait of a fictional bicycle courier waiting under a transparent umbrella at a rainy tram stop in Hamburg at blue hour. She wears a mustard raincoat, navy trousers, red sneakers, a silver helmet, and carries a teal messenger bag. Keep both hands visible, render the bicycle correctly, use realistic wet-street reflections, and include no readable brand names.

Generated through openrouter with model ID microsoft/mai-image-2.6. Archived settings were aspect ratio 1:1, provider default resolution, received 1024x1024, n=1, seed=provider random, actual cost $0.039287.

Open source

Open Infographic at original size

Infographic

This sample tests Numbers, labels, and visual organization.

Prompt and settings

Create a square German infographic titled SO FUNKTIONIERT PHOTOSYNTHESE. Show exactly four numbered steps in this order: 1 LICHT, 2 WASSER, 3 CO₂, 4 ZUCKER + SAUERSTOFF. Use a clean editorial science style with one plant cross-section, simple arrows, high contrast, legible labels, and no other text.

Generated through openrouter with model ID microsoft/mai-image-2.6. Archived settings were aspect ratio 1:1, provider default resolution, received 1024x1024, n=1, seed=provider random, actual cost $0.039317.

Open source

Open Product photography at original size

Product photography

This sample tests Materials, reflections, and fine details.

Prompt and settings

Create a square premium product photograph of a brushed titanium wristwatch standing upright on dark green marble. The watch has a cream dial, thin black hands set to 10:10, twelve distinct hour markers, a realistic crown, and a dark brown leather strap. Soft window light from the left, controlled reflections, shallow depth of field, no text, no logo, no extra objects.

Generated through openrouter with model ID microsoft/mai-image-2.6. Archived settings were aspect ratio 1:1, provider default resolution, received 1024x1024, n=1, seed=provider random, actual cost $0.039292.

Open source

Open Typography and layout at original size

Typography and layout

This sample tests Legible text, hierarchy, and composition.

Prompt and settings

Create a square editorial poster for an imaginary night train called MONDFALTER. Show the exact German headline MONDFALTER and the exact subline BERLIN NACH LISSABON. Use a restrained midnight-blue and warm-cream palette, one stylized moth, strong Swiss-grid typography, generous negative space, and no additional words or logos.

Generated through openrouter with model ID microsoft/mai-image-2.6. Archived settings were aspect ratio 1:1, provider default resolution, received 1024x1024, n=1, seed=provider random, actual cost $0.039252.

Open source

All benchmark results

Every published result with its source, test setup and comparison models.

Show all 41 published benchmark results

Published benchmarks

Text to image

Text to Image Arena

Elo ratings summarize pairwise comparisons within this leaderboard. They are not percentages. The scale does not express absolute quality or linear gaps between models.

GPT Image 2.5 Sunburst1,222.56 Elo

GPT Image 2.5 Flare1,207.17 Elo

GPT Image 21,188.55 Elo

Grok Imagine Image 2.01,167.96 Elo

MAI-Image-2.61,166.73 Elo

Nano Banana 21,140.04 Elo

Muse Image1,126.51 Elo

GPT Image 1.51,122.98 Elo

MAI-Image-2.6 Flash1,120.06 Elo

MAI-Image-2.51,118.46 Elo

Nano Banana Pro1,117.96 Elo

Nano Banana 2 Lite1,106.52 Elo

Seedream 5.0 Pro1,101.34 Elo

Grok Imagine Image Quality1,048.87 Elo

Qwen-Image-2.11,043.13 Elo

Krea 2 Large1,034.15 Elo

Qwen Image 2.0 Pro1,030.07 Elo

MAI-Image-2.5 Flash1,029.38 Elo

FLUX.2 Flex1,026.13 Elo

Recraft V4.1 Utility1,025.94 Elo

FLUX.2 Max1,020.74 Elo

Krea 2 Medium Turbo1,020.6 Elo

HiDream-O1-Image-1.51,017.32 Elo

Ideogram 4.51,016.15 Elo

Wan 2.6 Text to Image1,015.27 Elo

FLUX.2 Pro1,002.84 Elo

Ideogram 4.01,002.25 Elo

HunyuanImage 3.0 Instruct987.91 Elo

HiDream-O1-Image975.27 Elo

FLUX.2 Klein 9B929.74 Elo

higher is betterAs of November 3, 2026Scale: 0 Elo - 1,222.56 Elo

Source: Artificial Analysis

|

gradually.ai

Methodology and context

Metric: elo-rating
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 30
Methodology and context: AA appearances: 8395. This count does not establish an independent sample. Lower rank: 4. Upper rank: 6. Source model ID: 5d931c2e-797e-406e-8eb9-83ce18b5710c. Evidence capture ID: aa-image-text-to-image-v2.1-2026-11-03-b02bfcda6707. Source payload hash: b02bfcda6707b185b316c6b1c2e00ef9efe82010f1e2f2903d32096adb7e1cd9

Source: Artificial Analysis

LMArena Text to Image: 3D modeling

GPT Image 2.5 Sunburst1,416.06 points

GPT Image 2.5 Flare1,378.01 points

GPT Image 21,362.88 points

MAI-Image-2.61,325.39 points

Grok Imagine Image 2.01,291.64 points

Reve 2.11,280.92 points

Nano Banana 21,254.79 points

MAI-Image-2.51,251.32 points

Seedream 5.0 Pro1,247.62 points

Muse Image1,238.77 points

Nano Banana Pro1,232.98 points

Nano Banana 2 Lite1,225.57 points

GPT Image 1.51,217.58 points

Qwen-Image-2.11,213.34 points

FLUX.2 Flex1,172.95 points

FLUX.2 Max1,160.31 points

FLUX.2 Pro1,144.63 points

HiDream-O1-Image1,121.95 points

Krea 2 Large1,115.16 points

FLUX.2 Klein 9B1,078.47 points

Stable Diffusion 3.5 Large962.94 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,416.06 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 2,821
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Agentic image generation

GPT Image 2.5 Sunburst1,396.62 points

GPT Image 2.5 Flare1,358.45 points

GPT Image 21,338.59 points

MAI-Image-2.61,313.89 points

Grok Imagine Image 2.01,285.91 points

Nano Banana 21,256.3 points

Seedream 5.0 Pro1,241.22 points

GPT Image 1.51,236.53 points

Muse Image1,236.35 points

Nano Banana Pro1,234.68 points

Qwen-Image-2.11,230.55 points

Krea 2 Large1,143.91 points

FLUX.2 Pro1,136.63 points

HiDream-O1-Image1,110.85 points

GPT Image 2.5 Sunburst

GPT Image 2.5 Flare

GPT Image 2

MAI-Image-2.6

Grok Imagine Image 2.0

higher is betterAs of October 28, 2026Scale: 0 points - 1,396.62 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 14
Sample: 2,359
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Art

GPT Image 2.5 Sunburst1,408.7 points

GPT Image 2.5 Flare1,386.82 points

GPT Image 21,369.63 points

MAI-Image-2.61,337.16 points

Grok Imagine Image 2.01,326.14 points

Reve 2.11,290.95 points

MAI-Image-2.51,276.56 points

Seedream 5.0 Pro1,259.09 points

Muse Image1,257.69 points

Nano Banana 2 Lite1,255.76 points

Nano Banana 21,240.58 points

Qwen-Image-2.11,234.02 points

GPT Image 1.51,224.88 points

Nano Banana Pro1,223.81 points

FLUX.2 Max1,168.71 points

FLUX.2 Pro1,159.46 points

FLUX.2 Flex1,155.25 points

Krea 2 Large1,123.59 points

HiDream-O1-Image1,116.18 points

FLUX.2 Klein 9B1,078.57 points

Stable Diffusion 3.5 Large997.21 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,408.7 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 4,856
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Cartoon

GPT Image 2.5 Sunburst1,456.05 points

GPT Image 2.5 Flare1,430.72 points

GPT Image 21,402.35 points

Grok Imagine Image 2.01,352.28 points

MAI-Image-2.61,348.23 points

Reve 2.11,309.62 points

Muse Image1,275.92 points

MAI-Image-2.51,268.84 points

Seedream 5.0 Pro1,264.27 points

Nano Banana 21,259.12 points

Nano Banana 2 Lite1,256.28 points

GPT Image 1.51,243.84 points

Nano Banana Pro1,243.18 points

Qwen-Image-2.11,233.67 points

FLUX.2 Max1,167.46 points

FLUX.2 Flex1,163.21 points

FLUX.2 Pro1,158.68 points

Krea 2 Large1,119.32 points

HiDream-O1-Image1,118.17 points

FLUX.2 Klein 9B1,063.56 points

Stable Diffusion 3.5 Large945.19 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,456.05 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 13,098
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Commercial design

GPT Image 2.5 Sunburst1,418.65 points

GPT Image 2.5 Flare1,386.73 points

GPT Image 21,385.11 points

MAI-Image-2.61,331.62 points

Grok Imagine Image 2.01,325.93 points

Reve 2.11,315.28 points

Nano Banana 21,267.45 points

MAI-Image-2.51,259.29 points

Muse Image1,259.08 points

Nano Banana 2 Lite1,248.33 points

Nano Banana Pro1,245.72 points

GPT Image 1.51,237.8 points

Qwen-Image-2.11,237.63 points

Seedream 5.0 Pro1,235.73 points

FLUX.2 Flex1,172.42 points

FLUX.2 Max1,163.76 points

FLUX.2 Pro1,154.32 points

HiDream-O1-Image1,129.14 points

Krea 2 Large1,110.75 points

FLUX.2 Klein 9B1,070.23 points

Stable Diffusion 3.5 Large936.82 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,418.65 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 10,174
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Overall

GPT Image 2.5 Sunburst1,427.39 points

GPT Image 2.5 Flare1,401.27 points

GPT Image 21,381.28 points

MAI-Image-2.61,328.65 points

Grok Imagine Image 2.01,316.57 points

Reve 2.11,300.12 points

Muse Image1,268.83 points

Nano Banana 21,256.98 points

MAI-Image-2.51,253 points

Seedream 5.0 Pro1,250.44 points

Nano Banana 2 Lite1,249.44 points

Nano Banana Pro1,246.19 points

GPT Image 1.51,236.25 points

Qwen-Image-2.11,225.11 points

FLUX.2 Max1,161.4 points

FLUX.2 Flex1,156.21 points

FLUX.2 Pro1,152.7 points

Krea 2 Large1,115.21 points

HiDream-O1-Image1,114.76 points

FLUX.2 Klein 9B1,069.79 points

Stable Diffusion 3.5 Large938.33 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,427.39 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 34,596
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Photorealism

GPT Image 2.5 Sunburst1,432.65 points

GPT Image 2.5 Flare1,414.92 points

GPT Image 21,378.3 points

MAI-Image-2.61,329.8 points

Grok Imagine Image 2.01,311.03 points

Reve 2.11,301.89 points

Muse Image1,280.83 points

Seedream 5.0 Pro1,273.76 points

Nano Banana 21,261.86 points

Nano Banana Pro1,260.53 points

Nano Banana 2 Lite1,258.76 points

MAI-Image-2.51,250.53 points

GPT Image 1.51,246.82 points

Qwen-Image-2.11,226.53 points

FLUX.2 Max1,160.89 points

FLUX.2 Pro1,151.41 points

FLUX.2 Flex1,145.43 points

Krea 2 Large1,128.62 points

HiDream-O1-Image1,109.13 points

FLUX.2 Klein 9B1,090.95 points

Stable Diffusion 3.5 Large928.42 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,432.65 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 12,022
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Portraits

GPT Image 2.5 Sunburst1,461.04 points

GPT Image 2.5 Flare1,438.88 points

GPT Image 21,425.97 points

MAI-Image-2.61,338.94 points

Grok Imagine Image 2.01,331.53 points

Reve 2.11,316.28 points

Muse Image1,293.55 points

Seedream 5.0 Pro1,286.27 points

GPT Image 1.51,259.26 points

Nano Banana 21,256.64 points

Nano Banana Pro1,256.22 points

MAI-Image-2.51,254.66 points

Qwen-Image-2.11,251.64 points

Nano Banana 2 Lite1,244.36 points

FLUX.2 Max1,156.6 points

FLUX.2 Flex1,143.86 points

FLUX.2 Pro1,143.83 points

Krea 2 Large1,122.8 points

HiDream-O1-Image1,105.43 points

FLUX.2 Klein 9B1,091.81 points

Stable Diffusion 3.5 Large941.9 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,461.04 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 6,165
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

LMArena Text to Image: Text rendering

GPT Image 2.5 Sunburst1,465.68 points

GPT Image 2.5 Flare1,435.32 points

GPT Image 21,423.28 points

MAI-Image-2.61,364.74 points

Grok Imagine Image 2.01,349.28 points

Reve 2.11,339.17 points

Nano Banana 21,293.57 points

Muse Image1,293.25 points

MAI-Image-2.51,280.64 points

Nano Banana 2 Lite1,273.68 points

Nano Banana Pro1,270.84 points

Seedream 5.0 Pro1,254.68 points

GPT Image 1.51,251.49 points

Qwen-Image-2.11,250.45 points

FLUX.2 Flex1,171.52 points

FLUX.2 Max1,165.47 points

FLUX.2 Pro1,154.74 points

HiDream-O1-Image1,133.86 points

Krea 2 Large1,116.56 points

FLUX.2 Klein 9B1,057.75 points

Stable Diffusion 3.5 Large921.08 points

higher is betterAs of October 28, 2026Scale: 0 points - 1,465.68 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 21
Sample: 10,580
Methodology and context: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source: LMArena

Image editing

Image Editing Arena

Elo ratings summarize pairwise comparisons within this leaderboard. They are not percentages. The scale does not express absolute quality or linear gaps between models.

GPT Image 2.5 Sunburst1,199.48 Elo

GPT Image 2.5 Flare1,182.52 Elo

MAI-Image-2.61,150.73 Elo

MAI-Image-2.6 Flash1,144.89 Elo

GPT Image 21,136.54 Elo

Seedream 5.0 Pro1,129.33 Elo

Muse Image1,129.11 Elo

MAI-Image-2.51,126.67 Elo

Nano Banana 21,124.08 Elo

GPT Image 1.51,118.78 Elo

Grok Imagine Image 2.01,117.71 Elo

Nano Banana Pro1,108.9 Elo

Grok Imagine Image Quality1,096.22 Elo

MAI-Image-2.5 Flash1,090.4 Elo

Qwen-Image-2.11,081.38 Elo

Ideogram 4.51,068.66 Elo

HunyuanImage 3.0 Instruct1,064.33 Elo

Qwen Image 2.0 Pro1,058.03 Elo

Nano Banana 2 Lite1,055.27 Elo

FLUX.2 Klein 9B1,015.61 Elo

FLUX.2 Pro1,013.07 Elo

FLUX.2 Flex1,001.91 Elo

FLUX.2 Max998.74 Elo

HiDream-O1-Image954.61 Elo

higher is betterAs of November 3, 2026Scale: 0 Elo - 1,199.48 Elo

Source: Artificial Analysis

|

gradually.ai

Methodology and context

Metric: elo-rating
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 24
Methodology and context: AA appearances: 16713. This count does not establish an independent sample. Lower rank: 4. Upper rank: 7. Source model ID: 5d931c2e-797e-406e-8eb9-83ce18b5710c. Evidence capture ID: aa-image-image-editing-v2.1-2026-11-03-3c9ca7a083ad. Source payload hash: 3c9ca7a083ada0be4f50daced49cc4d1282114c6ba75a56557795b4ea75aa6f5

Source: Artificial Analysis

LMArena Image Editing: Agentic image generation

GPT Image 2.5 Sunburst1,528.73 points

GPT Image 2.5 Flare1,495.45 points

GPT Image 21,458.81 points

Grok Imagine Image 2.01,431.88 points

MAI-Image-2.61,416.16 points

Nano Banana Pro1,409.95 points

Seedream 5.0 Pro1,409.44 points

Muse Image1,393.39 points

Nano Banana 21,377.25 points

Qwen-Image-2.11,367.91 points

GPT Image 1.51,358.38 points

Ideogram 4.51,295.89 points

FLUX.2 Pro1,215.6 points

GPT Image 2.5 Sunburst

GPT Image 2.5 Flare

GPT Image 2

Grok Imagine Image 2.0

MAI-Image-2.6

higher is betterAs of October 28, 2026Scale: 0 points - 1,528.73 points

Source: LMArena

|

gradually.ai

Methodology and context

Metric: arena-score
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 13
Sample: 2,807
Methodology and context: Evidence capture ID: lmarena-image_edit-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-efe70ea24802. Source payload hash: efe70ea24802bd2e7db887a8fc6562f6c29ecdda5c404685383a91e28cc6f17d

Source: LMArena

Recorded measurement details

Text to Image Arena (Show measurement, test conditions, and source)

Source value: 1166.73
Score: 1166.73
Metric: elo-rating
Unit: Elo
Benchmark version: v2.1
Direction: Higher is better
Source rank: 6
Confidence interval: 12
Lower confidence bound: 1154.73
Upper confidence bound: 1178.73
Source type: independent-evaluator
Evaluator: Artificial Analysis
Retrieved at: 2026-11-03
Methodology: AA appearances: 8395. This count does not establish an independent sample. Lower rank: 4. Upper rank: 6. Source model ID: 5d931c2e-797e-406e-8eb9-83ce18b5710c. Evidence capture ID: aa-image-text-to-image-v2.1-2026-11-03-b02bfcda6707. Source payload hash: b02bfcda6707b185b316c6b1c2e00ef9efe82010f1e2f2903d32096adb7e1cd9
USD price per 1,000 tokens: 38.912

Source Artificial Analysis

Image Editing Arena (Show measurement, test conditions, and source)

Source value: 1150.73
Score: 1150.73
Metric: elo-rating
Unit: Elo
Benchmark version: v2.1
Direction: Higher is better
Source rank: 6
Confidence interval: 12
Lower confidence bound: 1138.73
Upper confidence bound: 1162.73
Source type: independent-evaluator
Evaluator: Artificial Analysis
Retrieved at: 2026-11-03
Methodology: AA appearances: 16713. This count does not establish an independent sample. Lower rank: 4. Upper rank: 7. Source model ID: 5d931c2e-797e-406e-8eb9-83ce18b5710c. Evidence capture ID: aa-image-image-editing-v2.1-2026-11-03-3c9ca7a083ad. Source payload hash: 3c9ca7a083ada0be4f50daced49cc4d1282114c6ba75a56557795b4ea75aa6f5
USD price per 1,000 tokens: 38.912

Source Artificial Analysis

LMArena Text to Image: 3D modeling (Show measurement, test conditions, and source)

Source value: 1325.3875978614046
Score: 1325.3875978614046
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 5
Sample: 2821
Confidence interval: 11.853953
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Agentic image generation (Show measurement, test conditions, and source)

Source value: 1313.8927024061827
Score: 1313.8927024061827
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 4
Sample: 2359
Confidence interval: 15.437993
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Art (Show measurement, test conditions, and source)

Source value: 1337.1640445169219
Score: 1337.1640445169219
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 4
Sample: 4856
Confidence interval: 9.59299
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Cartoon (Show measurement, test conditions, and source)

Source value: 1348.2308823920966
Score: 1348.2308823920966
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 5
Sample: 13098
Confidence interval: 7.269105
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Commercial design (Show measurement, test conditions, and source)

Source value: 1331.6184354347004
Score: 1331.6184354347004
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 4
Sample: 10174
Confidence interval: 7.674673
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Overall (Show measurement, test conditions, and source)

Source value: 1328.6451815813434
Score: 1328.6451815813434
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 4
Sample: 34596
Confidence interval: 5.380031
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Photorealism (Show measurement, test conditions, and source)

Source value: 1329.79568329776
Score: 1329.79568329776
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 5
Sample: 12022
Confidence interval: 7.422004
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Portraits (Show measurement, test conditions, and source)

Source value: 1338.942943482015
Score: 1338.942943482015
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 4
Sample: 6165
Confidence interval: 9.151663
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Text to Image: Text rendering (Show measurement, test conditions, and source)

Source value: 1364.743687664083
Score: 1364.743687664083
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 4
Sample: 10580
Confidence interval: 7.779766
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-text_to_image-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-32972b5ffdc5. Source payload hash: 32972b5ffdc597e97f4dc7e2bad6b89fd6020216e049e066a6dbc55f74233037

Source LMArena

LMArena Image Editing: Agentic image generation (Show measurement, test conditions, and source)

Source value: 1416.15967495405
Score: 1416.15967495405
Metric: arena-score
Unit: score
Direction: Higher is better
Source rank: 5
Sample: 2807
Confidence interval: 14.35863
Source type: official-leaderboard
Evaluator: LMArena
Retrieved at: 2026-10-28
Methodology: Evidence capture ID: lmarena-image_edit-6852ae7aa7c105ff8cf86bb1551d1b1fd52f9dd3-efe70ea24802. Source payload hash: efe70ea24802bd2e7db887a8fc6562f6c29ecdda5c404685383a91e28cc6f17d

Source LMArena

Page 1 of 4

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

The 100 most popular models of this category, 0 to 100, higher is more popular

20 of 100 models

Stable Diffusion XL 1.072

Anima Base v1.068

SDXL Turbo63

Stable Diffusion Turbo63

MAI-Image-2.645

PixArt-Sigma-XL-2-1024-MS44

Janus-Pro-7B44

HunyuanImage 3.0 Instruct44

MOSS-Image-Base44

HiDream-O1-Image43

Stable Diffusion XL 1.0

Anima Base v1.0

SDXL Turbo

Stable Diffusion Turbo

MAI-Image-2.6

PixArt-Sigma-XL-2-1024-MS

Janus-Pro-7B

HunyuanImage 3.0 Instruct

MOSS-Image-Base

HiDream-O1-Image

As of November 2026 · Source: gradually.ai popularity index

Over time

Weekly values from October 18, 2026 to November 5, 2026

MAI-Image-2.6FLUX.2 Klein 9BHunyuanImage 3.0 InstructGPT Image 2FLUX.2 Flex

MAI-Image-2.6 compared

Compare with another model

First image modelSecond image model

Predecessor and successor

No predecessor recorded for MAI-Image-2.6
MAI-Image-2.6 has no successor yet

The most popular comparisons

Show all 139 comparisons

All specifications

Input and output

Context window: 32,000 tokens; no source
Maximum output: 2,359,296 pixels; Source: Maximum output

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking: Unknown; no source
Watermarking technique: Not known; no source

Licence and access

Weights: Proprietary; Source: Weights
Access: API; Source: Access
Lifecycle: Preview in Microsoft Foundry; Source: Lifecycle

Further details

Image capabilities: Text-to-image generation and precise image-to-image editing; Source: Image capabilities
Editing controls: Target objects, layouts, and text, and clean up motion blur; Source: Editing controls
Output format and size: PNG, at least 768 pixels on each side and at most 2,359,296 pixels in total; Source: Output format and size
Aspect ratio: Can choose the aspect ratio from the prompt and image inputs; Source: Aspect ratio
Web grounding: Can use current Bing Search results as additional image context; Source: Web grounding
API authentication: Microsoft Entra ID or an API key; Source: API authentication

Prices

Representative price: $0.038912 per imageSource
OpenRouter (microsoft/mai-image-2.6): $0.06 per image (estimated)Source

Frequently asked questions about MAI-Image-2.6

How good is MAI-Image-2.6?

Rank 4 of 21 in the arena rating. Rank 28 of 108 in our own test.

Arena rating from Oct 28, 2026 · GraduallyBench from Nov 3, 2026

Arena ratingfrom user votes · higher is better

GPT Image 2.5 Sunburst1,427

GPT Image 2.5 Flare1,401

GPT Image 21,381

MAI-Image-2.6Rank 4 of 211,329

Grok Imagine Image 2.01,317

Reve 2.11,300

Muse Image1,269

GraduallyBenchour own rating, 0 to 100 · higher is better

ChatGPT Image Latest (API-Alias)69

MAI-Image-2.569

FLUX.2 Max69

MAI-Image-2.6Rank 28 of 10868

GPT Image 1.568

Qwen Image 3.068

MAI-Image-2.6 Flash67

What is MAI-Image-2.6 good at?

Best at Product photography (88 out of 100), weakest at Infographic (61).

GraduallyBench, one image per task

Typography and layoutLegible text, hierarchy, and composition62out of 100Rank 45 of 108 in this task
Product photographyMaterials, reflections, and fine details88out of 100Rank 7 of 108 in this task
Character and detailAnatomy, clothing, and spatial consistency63out of 100Rank 34 of 108 in this task
InfographicNumbers, labels, and visual organization61out of 100Rank 51 of 108 in this task

How much does an image with MAI-Image-2.6 cost?

About $0.039 per image. That is fairly cheap: a typical image model costs $0.06.

Price per imageDollars per image · lower is better

FLUX.2 Pro$0.03

Nano Banana 2 Lite$0.034

Recraft V4.1 Utility$0.035

MAI-Image-2.6Rank 10 of 28$0.039

HiDream-O1-Image$0.04

MAI-Image-2.5$0.048

Grok Imagine Image Quality$0.05

Sources and data date

Show all 24 sources

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

Evaluator	Test series	Composite indices	Recorded active measurements
LMArena	0	0	1

Without vendor claims and historical or disputed results; those stay labelled in the single values. 1 measurements are not yet assigned to a test series.

Type	Evidence and data date
Research status	Research date October 17, 2026. 11 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not published	exact-serving-revision Note The evaluator shows the exact product row but does not disclose an API route or serving checkpoint revision.
Additional source	Microsoft AI (retrieved September 24, 2026) · Microsoft Foundry MAI-Image-2.6 catalog · Editorial description reviewed October 2, 2026
Additional source	Image pricing (retrieved October 17, 2026; October 28, 2026; November 3, 2026)
Additional source	Lifecycle data · Microsoft Foundry July and August 2026 updates · Editorial description reviewed October 2, 2026
Additional source	Microsoft AI (retrieved September 24, 2026) · Microsoft Foundry MAI image deployment guide · Editorial description reviewed October 2, 2026
Additional source	OpenRouter
Additional source	Artificial Analysis (retrieved October 17, 2026; October 23, 2026; October 28, 2026; November 3, 2026)
Additional source	LMArena (retrieved October 28, 2026)1 2
Additional source	Gradually-Bildtest (retrieved November 3, 2026)
Additional source	Arena (retrieved October 19, 2026)
Additional source	Quantslant (retrieved October 19, 2026)
Additional source	Arena.ai Text-to-Image leaderboard (retrieved October 21, 2026; October 23, 2026)
Additional source	Microsoft AI (retrieved October 22, 2026)
Additional source	DesignArena (retrieved October 23, 2026; October 24, 2026; October 28, 2026)
Additional source	Model metadata (retrieved October 17, 2026)
Additional source	Model metadata (retrieved October 23, 2026)1 2
Additional source	Model metadata (retrieved October 28, 2026)
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Ollama: cumulative downloads for the popularity index
Additional source	Replicate: cumulative runs for the popularity index
Additional source	Google Trending Now: limited feed selection for Germany and the US
Additional source	Creative Commons Attribution 4.0