MOSS-Image-Basevs.HunyuanImage 3.0

Which one is better?

In our own test, MOSS-Image-Base leads. Neither is in the arena. Neither has an API price.

Test images

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

MOSS-Image-Base: Typography and layout, first image from the fixed test run — MOSS-Image-Base51 out of 100

HunyuanImage 3.0: Typography and layout, first image from the fixed test run — HunyuanImage 3.013 out of 100

Result for “Typography and layout”:MOSS-Image-Base 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.

Character and detail

MOSS-Image-Base: manual, local PyTorch 2.13.0, Diffusers 0.41.0; OpenMOSS-Team/MOSS-Image-Base@85975793d8b3d432f1b6e27f352abaa4c2a4c0d3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.0, seed=921965437095743280; mps for text encoder and denoising; cpu for native VAE decode; bfloat16 for text encoder and transformer; float32 for VAE; unconditional skip layers=[22]; CPU operator fallback permitted, actual fallback use was not measured; official experimental MPS compatibility runtime OpenMOSS/MOSS-Image@1961ca3d111b828e65278656cde0c8f94484f4dd, generated Nov 2, 2026, 10:39 PM. Source

HunyuanImage 3.0: fal, requested 1024x1024, n=1, seed=896514467, prompt expansion disabled, Estimate reserves 2 MP at Fal's $0.10 per MP rate, estimated cost $0.200000, generated Nov 2, 2026, 3:08 PM. Source

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

MOSS-Image-Base image SHA 256: 313d95f3e7766e861d46b0563c81bb2a02782708004c24c7aab0c46e8587ff83

HunyuanImage 3.0 image SHA 256: 872ff28241f5b37bfd5163d98d5e39abdd365856a3689bc375df1eacf4532598

Infographic

MOSS-Image-Base: manual, local PyTorch 2.13.0, Diffusers 0.41.0; OpenMOSS-Team/MOSS-Image-Base@85975793d8b3d432f1b6e27f352abaa4c2a4c0d3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.0, seed=2567672344276262662; mps for text encoder and denoising; cpu for native VAE decode; bfloat16 for text encoder and transformer; float32 for VAE; unconditional skip layers=[22]; CPU operator fallback permitted, actual fallback use was not measured; official experimental MPS compatibility runtime OpenMOSS/MOSS-Image@1961ca3d111b828e65278656cde0c8f94484f4dd, generated Nov 2, 2026, 10:43 PM. Source

HunyuanImage 3.0: fal, requested 1024x1024, n=1, seed=1046110920, prompt expansion disabled, Estimate reserves 2 MP at Fal's $0.10 per MP rate, estimated cost $0.200000, generated Nov 2, 2026, 3:09 PM. Source

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

MOSS-Image-Base image SHA 256: 3175582007cf377490ab51f73a57ebddb312f523ac5bfb54ae7598e72eb3edca

HunyuanImage 3.0 image SHA 256: 957c37f6232bdfed3ba5909778dbf599d69c74b1a0ebb998f3d437018007ee3a

Product photography

MOSS-Image-Base: manual, local PyTorch 2.13.0, Diffusers 0.41.0; OpenMOSS-Team/MOSS-Image-Base@85975793d8b3d432f1b6e27f352abaa4c2a4c0d3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.0, seed=8912274556852986311; mps for text encoder and denoising; cpu for native VAE decode; bfloat16 for text encoder and transformer; float32 for VAE; unconditional skip layers=[22]; CPU operator fallback permitted, actual fallback use was not measured; official experimental MPS compatibility runtime OpenMOSS/MOSS-Image@1961ca3d111b828e65278656cde0c8f94484f4dd, generated Nov 2, 2026, 10:35 PM. Source

HunyuanImage 3.0: fal, requested 1024x1024, n=1, seed=1902047757, prompt expansion disabled, Estimate reserves 2 MP at Fal's $0.10 per MP rate, estimated cost $0.200000, generated Nov 2, 2026, 3:08 PM. Source

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

MOSS-Image-Base image SHA 256: af9ea47eb1554bcaaaadf8860fecfa5bb69095a58bf346e3d9a73be8240a2a7c

HunyuanImage 3.0 image SHA 256: a7847d09d9ae2c2fa2c5311a5ff4a6e9357e24736c374f14af38ed3423f88721

Typography and layout

MOSS-Image-Base: manual, local PyTorch 2.13.0, Diffusers 0.41.0; OpenMOSS-Team/MOSS-Image-Base@85975793d8b3d432f1b6e27f352abaa4c2a4c0d3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.0, seed=9219943416610765538; mps for text encoder and denoising; cpu for native VAE decode; bfloat16 for text encoder and transformer; float32 for VAE; unconditional skip layers=[22]; CPU operator fallback permitted, actual fallback use was not measured; official experimental MPS compatibility runtime OpenMOSS/MOSS-Image@1961ca3d111b828e65278656cde0c8f94484f4dd, generated Nov 2, 2026, 10:31 PM. Source

HunyuanImage 3.0: fal, requested 1024x1024, n=1, seed=1922646419, prompt expansion disabled, Estimate reserves 2 MP at Fal's $0.10 per MP rate, estimated cost $0.200000, generated Nov 2, 2026, 3:08 PM. Source

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

MOSS-Image-Base image SHA 256: e87b1d9d7aa09487c6b9be505746dacb2f3be551f29a5d85424a44a2a88089c5

HunyuanImage 3.0 image SHA 256: 2ad121402920a53918c6a31f30a9e93f9e0dc5cac94545edd78454f96bd36996

The numbers at a glance

Arena ratingfrom votes by users · higher is better

n/a

MOSS-Image-Base

not in the arena

n/a

HunyuanImage 3.0

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better

MOSS-Image-Base

Rank 61 of 108

HunyuanImage 3.0

Rank 89 of 108

Price per imagedollars through the API · lower is better

n/a

MOSS-Image-Base

no API price

n/a

HunyuanImage 3.0

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.

MOSS-Image-Base

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

HunyuanImage 3.0

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.

GraduallyBench from Nov 3, 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	MOSS-Image-Base	HunyuanImage 3.0
Ratings
Arena ratingfrom votes by users, higher is better	n/a	n/a
Arena rank	n/a	n/a
Arena votesmore votes, a safer rating	n/a	n/a
GraduallyBenchour own rating, 0 to 100	51	27
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition	51	13
Product photographyMaterials, reflections, and fine details	36	34
Character and detailAnatomy, clothing, and spatial consistency	46	24
InfographicNumbers, labels, and visual organization	71	36
API price
Per imagelower is better	n/a	n/a
100 images	n/a	n/a
Profile
Released	Oct 16, 2026	Sep 28, 2025
Weightsopen means: runs locally	open	open
Access	local	local
Parameters	5 billion	80 billion
Largest output	not consistently published	not consistently published
Plans in Germanynaming exactly this model	n/a	n/a

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

Sources: MOSS-Image-Base model source (verified Nov 1, 2026) · HunyuanImage 3.0 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 MOSS-Image-Base and HunyuanImage 3.0 from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare MOSS-Image-Base and HunyuanImage 3.0 with the models people compare them with most often.

Compare another matchup

First image modelSecond image model