Skip to main content

AlibabaInclusionAIImage model matchup

Qwen Image 2.0 Provs.LLaDA-Image

Which one is better?

In our own benchmark GraduallyBench Image v2, Qwen Image 2.0 Pro leads. Neither is in the arena. LLaDA-Image has no API price.

GraduallyBench Image v2Our own benchmark0 to 100 · higher is better
76

Qwen Image 2.0 Pro

Rank 18 of 108

41

LLaDA-Image

Rank 65 of 108

Price per imagedollars through the API · lower is better
$0.075

Qwen Image 2.0 Pro

n/a

LLaDA-Image

Ranks among the 21 image models with an arena rating and the 108 with a GraduallyBench Image v2 run; prices compared with 28 models with an API price.

Test images from GraduallyBench Image v2

Our own benchmarkThe images come from GraduallyBench Image v2, our own benchmark: every model gets the same four tasks, and two AI judges compare the images in pairs and blind.

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

Qwen Image 2.0 Pro: Infographic, first image from the fixed test run
Qwen Image 2.0 Pro93 out of 100
LLaDA-Image: Infographic, first image from the fixed test run
LLaDA-Image29 out of 100

Result for “Infographic”:Qwen Image 2.0 Pro ahead

Prompt given to both modelsCreate 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.

What the rating looks at: Numbers, labels, and visual organization.

“Ahead”: the 95 % ranges of both values do not overlap. “Slightly ahead”: a higher value, but the ranges overlap.

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.

Character and detail

Qwen Image 2.0 Pro: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:21 PM. Source

LLaDA-Image: manual, 2.13.0 PyTorch, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image@e4e2703f410f7ddb6ee8d6b09dac6a8ec5093039; native 1024x1024, 1:1; one request, first output; steps=50, guidance=5.0; seed=6428048197886116697; MPS generator; BF16 text/diffusion components, FP32 denoising latent, receipt-reported CPU FP32 native VAE decode; negative_prompt argument omitted (None default, not the manufacturer's empty-string example); MPS community PR @ 5395ce055da9975b09f71e21b865973d409bb3cd; official pinned text encoder uses eager MoE fallback and SDPA attention; other image-component attention paths were not independently captured by this review; no CUDA numerical equivalence claimed; MPS fallback permitted, actual fallback operations not measured; two submissions: the original request stopped at the 60-GiB free-space safeguard before image or latent (supervisor exit 143); one technical same-request/same-seed retry produced the retained first completed output; no quality reroll, generated Nov 3, 2026, 8:50 AM. Source

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

Qwen Image 2.0 Pro image SHA 256: 05f19da3e2d0eedb17d098c1b21abf8c83c663afa59083f0ce7a67a948e8a263

LLaDA-Image image SHA 256: a34d96894f5500c5bce2c16cca14ae09748f3aff4d47d15f8aff0d1be5749211

Infographic

Qwen Image 2.0 Pro: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:22 PM. Source

LLaDA-Image: manual, 2.13.0 PyTorch, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image@e4e2703f410f7ddb6ee8d6b09dac6a8ec5093039; native 1024x1024, 1:1; one request, first output; steps=50, guidance=5.0; seed=1354677768900879717; MPS generator; BF16 text/diffusion components, FP32 denoising latent, receipt-reported CPU FP32 native VAE decode; negative_prompt argument omitted (None default, not the manufacturer's empty-string example); MPS community PR @ 5395ce055da9975b09f71e21b865973d409bb3cd; official pinned text encoder uses eager MoE fallback and SDPA attention; other image-component attention paths were not independently captured by this review; no CUDA numerical equivalence claimed; MPS fallback permitted, actual fallback operations not measured, generated Nov 3, 2026, 8:56 AM. Source

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

Qwen Image 2.0 Pro image SHA 256: 003fbe859de5df62a4f49e3d590c00f5036e208149d81226d1835871fd1e0410

LLaDA-Image image SHA 256: cad67ad3f1a475d12bc2152840dc98657f81b5c7318c09a9e9fa40fbb433372f

Product photography

Qwen Image 2.0 Pro: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:21 PM. Source

LLaDA-Image: manual, 2.13.0 PyTorch, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image@e4e2703f410f7ddb6ee8d6b09dac6a8ec5093039; native 1024x1024, 1:1; one request, first output; steps=50, guidance=5.0; seed=152405017095880227; MPS generator; BF16 text/diffusion components, FP32 denoising latent, receipt-reported CPU FP32 native VAE decode; negative_prompt argument omitted (None default, not the manufacturer's empty-string example); MPS community PR @ 5395ce055da9975b09f71e21b865973d409bb3cd; official pinned text encoder uses eager MoE fallback and SDPA attention; other image-component attention paths were not independently captured by this review; no CUDA numerical equivalence claimed; MPS fallback permitted, actual fallback operations not measured, generated Nov 3, 2026, 1:02 AM. Source

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

Qwen Image 2.0 Pro image SHA 256: b811a440655cae766c039dfc8f9efb1756be5dce63ef22cd95deb1fe7dab6a3a

LLaDA-Image image SHA 256: 888e173c25c26ff1dcbfd07aad01ffc6a71cc7735099d3e943e60dfddfe31553

Typography and layout

Qwen Image 2.0 Pro: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 2:00 PM. Source

LLaDA-Image: manual, 2.13.0 PyTorch, Diffusers 0.39.0, Transformers 4.57.6; inclusionAI/LLaDA-Image@e4e2703f410f7ddb6ee8d6b09dac6a8ec5093039; native 1024x1024, 1:1; one request, first output; steps=50, guidance=5.0; seed=8778315039365154420; MPS generator; BF16 text/diffusion components, FP32 denoising latent, receipt-reported CPU FP32 native VAE decode; negative_prompt argument omitted (None default, not the manufacturer's empty-string example); MPS community PR @ 5395ce055da9975b09f71e21b865973d409bb3cd; official pinned text encoder uses eager MoE fallback and SDPA attention; other image-component attention paths were not independently captured by this review; no CUDA numerical equivalence claimed; MPS fallback permitted, actual fallback operations not measured, generated Nov 3, 2026, 12:40 AM. Source

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

Qwen Image 2.0 Pro image SHA 256: 4eb4915afa71f6a8c76979dcecadd70beabd3869f11ca092a725fd7b6c87b4f2

LLaDA-Image image SHA 256: 256a88541c9821a6242bb434fb520d4d6f0793c934c66acd1fd2e8edd8086b5d

Subscription and API

Qwen Image 2.0 Pro

Subscription
At Alibaba only in team or business plans, from $20 a month per seat (Token Plan Team Team Standard Seat). 3 plans in Germany.
Via API
$0.075 per image, 100 images $7.50.
Weights
Closed weights.

LLaDA-Image

Subscription
n/a No plan in Germany names exactly this model.
Via API
No API price known (access: local).
Weights
Open weights.

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

GraduallyBench Image v2 from Nov 7, 2026 · API prices from Nov 10, 2026 · Subscription database from Sep 29, 2026 (44 days old)

All data side by side

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

KennzahlQwen Image 2.0 ProLLaDA-Image
Ratings
Arena ratingfrom votes by users, higher is bettern/an/a
Arena rankn/an/a
Arena votesmore votes, a safer ratingn/an/a
GraduallyBench Image v2our own benchmark, 0 to 1007641
GraduallyBench Image v2 by task
Typography and layoutLegible text, hierarchy, and composition7164
Product photographyMaterials, reflections, and fine details5739
Character and detailAnatomy, clothing, and spatial consistency8133
InfographicNumbers, labels, and visual organization9329
API price
Per imagelower is better$0.075n/a
100 images$7.50n/a
Profile
ReleasedApril 2026September 2026
Weightsopen means: runs locallyproprietaryopen
AccessAPIlocal
Parametersnot published6 billion
Largest output2,048 × 2,048 pixelsnot consistently published
Plans in Germanynaming exactly this model3n/a

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

Sources: Qwen Image 2.0 Pro model source (verified Aug 15, 2026) · LLaDA-Image model source (verified Nov 1, 2026)

Arena ratings by category

From blind votes on Artificial Analysis and LMArena; the sources are named under the card. “Ahead” means the lead is clear.

No shared arena score in the stored snapshot

The stored benchmark database contains no measurement for Qwen Image 2.0 Pro and LLaDA-Image from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare Qwen Image 2.0 Pro and LLaDA-Image with the models people compare them with most often.

Compare another matchup