MAI-Image-2.5vs.LLaDA-Image-Turbo

Which one is better?

In our own test, MAI-Image-2.5 leads. LLaDA-Image-Turbo has no arena rating yet. LLaDA-Image-Turbo has no API price.

Test images

Same task, the first image of each. The task with the biggest difference is preselected.

MAI-Image-2.5: Typography and layout, first image from the fixed test run — MAI-Image-2.582 out of 100

LLaDA-Image-Turbo: Typography and layout, first image from the fixed test run — LLaDA-Image-Turbo8 out of 100

Result for “Typography and layout”:MAI-Image-2.5 ahead

Prompt given to both modelsCreate 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.

What the rating looks at: Legible text, hierarchy, and composition.

“Ahead”: better in all three rating rounds. “Slightly ahead”: a higher value, but not in every round.

Test protocol of the images

Each test attempt uses the same archived prompt and one request. The first output of that attempt counts. Repeats after technical test failures are disclosed for the affected images. Files are normalized to 1,024 × 1,024 pixels without cropping. Provider resolution, randomness, quality tier and prompt processing can still differ. The images are a practical visual test, not a reconstruction of the arena configuration.

Typography and layout

MAI-Image-2.5: openrouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048473, generated Aug 14, 2026, 2:03 PM. Source

LLaDA-Image-Turbo: manual, 1:1, native 1024x1024, one request in the successful technical repeat, first output; two total submissions after a local download failure; seed 0 (fixed provider default, randomize_seed=false); BF16 transformer, FP8 text encoder; 4 steps, CFG 1.0; HF Space commit 7b02431121944d65b8065873b48eec736a0c1022; source PNG SHA-256 d1866b4994a664728de07f1c1897776bbf04409b4a00ad7fa98666036a78816d, generated Nov 2, 2026, 3:03 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: 2e012c3d14ee5dd02203a94f394ca80a5038cdebd6032b25a8b8a221f2255419

MAI-Image-2.5 image SHA 256: 4c8a60555b57b8c3b5ebaa131371d1bba9d31bfafe7aaa772554969f7c28cb08

LLaDA-Image-Turbo image SHA 256: de0c780506575a32050ce9ca3d3c4dd1a0c4cc7ccf7f7f2f6aaf17ad01295901

Product photography

MAI-Image-2.5: openrouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048513, generated Aug 14, 2026, 1:25 PM. Source

LLaDA-Image-Turbo: manual, PyTorch 2.13.0, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image-Turbo@f4afc52d925bbac4e22a1c947111fc1f127e37e5; native 1024x1024, 1:1; one request per local attempt, first output; steps=4, guidance=1.0; seed=0; separate MPS initial-latent generator and global MPS scheduler RNG streams with the same integer seed; original Turbo scheduler, stochastic_sampling=true, uniform sigmas, shift=3; BF16 text encoder and diffusion components, FP32 denoising latent and CPU FP32 VAE decode; historical poster uses FP8 text encoder and remains unchanged; no CUDA equivalence claimed; MPS fallback permitted, actual fallback operations not measured; historical Hugging Face Space product-photo attempt used the same archived prompt and seed 0 but a different request body and backend; its AppError left the output unresolved; this is not an exact request replay or a numerical-equivalence comparison, generated Nov 3, 2026, 9:48 AM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: c67fd71e91304458a2feef5c91efa8677260fab0c0099a29a2c52cb579f5b2f3

MAI-Image-2.5 image SHA 256: 67d4930b711fac61482231c4bc270cfcdb41c1b0daf653045781fb2126e237dd

LLaDA-Image-Turbo image SHA 256: e61fb35e0c12941d28d32b0e0125b934bc0857fc49216a7945406e3bf58b4316

Character and detail

MAI-Image-2.5: openrouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048508, generated Aug 14, 2026, 1:26 PM. Source

LLaDA-Image-Turbo: manual, PyTorch 2.13.0, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image-Turbo@f4afc52d925bbac4e22a1c947111fc1f127e37e5; native 1024x1024, 1:1; one request per local attempt, first output; steps=4, guidance=1.0; seed=7961710876241130415; separate MPS initial-latent generator and global MPS scheduler RNG streams with the same integer seed; original Turbo scheduler, stochastic_sampling=true, uniform sigmas, shift=3; BF16 text encoder and diffusion components, FP32 denoising latent and CPU FP32 VAE decode; historical poster uses FP8 text encoder and remains unchanged; no CUDA equivalence claimed; MPS fallback permitted, actual fallback operations not measured, generated Nov 3, 2026, 9:54 AM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: 985a75b38642a9f0e5e0c0e0c67f1b7f67780d42bd2168fa24e9a7a7a94b0da6

MAI-Image-2.5 image SHA 256: ca1c58688ae4601975d7e1f7b58bb43a0a999e736f610d9f51cd11e06284bced

LLaDA-Image-Turbo image SHA 256: aac02152ef0efc45749b3d0a626e9f4114418394aeb6a92d492651c564f40fee

Infographic

MAI-Image-2.5: openrouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048538, generated Aug 14, 2026, 1:26 PM. Source

LLaDA-Image-Turbo: manual, PyTorch 2.13.0, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image-Turbo@f4afc52d925bbac4e22a1c947111fc1f127e37e5; native 1024x1024, 1:1; one request per local attempt, first output; steps=4, guidance=1.0; seed=5156250474893981751; separate MPS initial-latent generator and global MPS scheduler RNG streams with the same integer seed; original Turbo scheduler, stochastic_sampling=true, uniform sigmas, shift=3; BF16 text encoder and diffusion components, FP32 denoising latent and CPU FP32 VAE decode; historical poster uses FP8 text encoder and remains unchanged; no CUDA equivalence claimed; MPS fallback permitted, actual fallback operations not measured, generated Nov 3, 2026, 9:56 AM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: ed65cb4e6a51e2336f8b149ad4cbb16e020f07d527ad94362e719d2dcf91d94f

MAI-Image-2.5 image SHA 256: eaf00a906231e8859f98ef5dc2cce7e8017797819aea453305a4ef0e970d4dbf

LLaDA-Image-Turbo image SHA 256: 4e8939998310be86b1ad3dc0744a5e3ceee95a00a9113ee6732f9c06cb6b3618

The numbers at a glance

Arena ratingfrom votes by users · higher is better

1,253

MAI-Image-2.5

Rank 9 of 21

n/a

LLaDA-Image-Turbo

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better

MAI-Image-2.5

Rank 26 of 108

LLaDA-Image-Turbo

Rank 83 of 108

Price per imagedollars through the API · lower is better

$0.048

MAI-Image-2.5

fairly cheap

n/a

LLaDA-Image-Turbo

no API price

Ranks among the 21 image models with an arena rating and the 108 with a GraduallyBench run; prices compared with 28 models with an API price.

MAI-Image-2.5

Subscription: No Microsoft AI subscription includes this model. Through other providers from €19.04 a month (Magnific Premium, with limited credits). 6 plans in Germany.
Via API: $0.048 per image, 100 images $4.81.

LLaDA-Image-Turbo

Subscription: n/a No plan in Germany names exactly this model.
Via API: No API price known. Access: local.
License: Open weights.

Subscriptions only count plans that name exactly this model. Subscription prices include VAT, API prices are in dollars.

Arena rating (LMArena) from Nov 6, 2026 · GraduallyBench from Nov 3, 2026 · API prices from Nov 6, 2026 · Subscription database from Sep 29, 2026 (38 days old)

All data side by side

The better value is bold. For ratings only when the gap is clear.

Kennzahl	MAI-Image-2.5	LLaDA-Image-Turbo
Ratings
Arena ratingfrom votes by users, higher is better	1,253	n/a
Arena rank	Rank 9 of 21	n/a
Arena votesmore votes, a safer rating	61,222	n/a
GraduallyBenchour own rating, 0 to 100	69	31
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition	82	8
Product photographyMaterials, reflections, and fine details	70	27
Character and detailAnatomy, clothing, and spatial consistency	55	53
InfographicNumbers, labels, and visual organization	69	36
API price
Per imagelower is better	$0.048	n/a
100 images	$4.81	n/a
Profile
Released	June 2026	Sep 4, 2026
Weightsopen means: runs locally	proprietary	open
Lifecycle	Preview in Microsoft Foundry	no notice
Access	API, Web app	local
Parameters	20 billion	6 billion
Largest output	1,048,576 pixels	not consistently published
Plans in Germanynaming exactly this model	6	n/a

API prices for one image at 1,024 × 1,024 pixels according to Artificial Analysis.

Sources: MAI-Image-2.5 model source, status source (verified Aug 15, 2026) · LLaDA-Image-Turbo model source (verified Nov 1, 2026)

Arena ratings by category

From blind votes on LMArena. “Ahead” means the lead is clear.

No shared arena score in the stored snapshot

The stored benchmark database contains no measurement for MAI-Image-2.5 and LLaDA-Image-Turbo from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare MAI-Image-2.5 and LLaDA-Image-Turbo with the models people compare them with most often.

Compare another matchup

First image modelSecond image model

The most popular comparisons

1MAI-Image-2.5 vs. Nano Banana 2.1