PixArt-Sigma-XL-2-1024-MSvs.HunyuanImage 3.0

Which one is better?

In our own test both are on par. 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.

PixArt-Sigma-XL-2-1024-MS: Infographic, first image from the fixed test run — PixArt-Sigma-XL-2-1024-MS5 out of 100

HunyuanImage 3.0: Infographic, first image from the fixed test run — HunyuanImage 3.036 out of 100

Result for “Infographic”:HunyuanImage 3.0 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”: 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

PixArt-Sigma-XL-2-1024-MS: manual, local PyTorch 2.13.0, Diffusers 0.36.0; PixArt-alpha/PixArt-Sigma-XL-2-1024-MS@e102b3591cc82e97071b8b4cb90d834d0c487207; native 1024x1024, 1:1; one request, first output; steps=20, guidance=4.5, seed=7226279879395584696; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE; native prompt preprocessing lowercases input, generated Nov 2, 2026, 5:58 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

PixArt-Sigma-XL-2-1024-MS image SHA 256: bde9f05db045ee4e3776436ec99b67b4ab2b1e1d98e7f3b9b389116880584dfa

HunyuanImage 3.0 image SHA 256: 872ff28241f5b37bfd5163d98d5e39abdd365856a3689bc375df1eacf4532598

Infographic

PixArt-Sigma-XL-2-1024-MS: manual, local PyTorch 2.13.0, Diffusers 0.36.0; PixArt-alpha/PixArt-Sigma-XL-2-1024-MS@e102b3591cc82e97071b8b4cb90d834d0c487207; native 1024x1024, 1:1; one request, first output; steps=20, guidance=4.5, seed=3371899936485956023; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE; native prompt preprocessing lowercases input, generated Nov 2, 2026, 5:59 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

PixArt-Sigma-XL-2-1024-MS image SHA 256: a1a6e8b2545200a59deb0cef3614e7421a91014fcb31e4e2725cc40ca1730017

HunyuanImage 3.0 image SHA 256: 957c37f6232bdfed3ba5909778dbf599d69c74b1a0ebb998f3d437018007ee3a

Product photography

PixArt-Sigma-XL-2-1024-MS: manual, local PyTorch 2.13.0, Diffusers 0.36.0; PixArt-alpha/PixArt-Sigma-XL-2-1024-MS@e102b3591cc82e97071b8b4cb90d834d0c487207; native 1024x1024, 1:1; one request, first output; steps=20, guidance=4.5, seed=2926417222682566089; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE; native prompt preprocessing lowercases input, generated Nov 2, 2026, 5:58 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

PixArt-Sigma-XL-2-1024-MS image SHA 256: 2012912767607d877602f52cd6766af87520f70db960cda065baa5282c377b51

HunyuanImage 3.0 image SHA 256: a7847d09d9ae2c2fa2c5311a5ff4a6e9357e24736c374f14af38ed3423f88721

Typography and layout

PixArt-Sigma-XL-2-1024-MS: manual, local PyTorch 2.13.0, Diffusers 0.36.0; PixArt-alpha/PixArt-Sigma-XL-2-1024-MS@e102b3591cc82e97071b8b4cb90d834d0c487207; native 1024x1024, 1:1; one request, first output; steps=20, guidance=4.5, seed=154650094805927878; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE; native prompt preprocessing lowercases input, generated Nov 2, 2026, 5:57 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

PixArt-Sigma-XL-2-1024-MS image SHA 256: 07f61240d9329bf99f8a3d21b7b437c65c4eeb1780c8f1164e0c8baf2a0f7085

HunyuanImage 3.0 image SHA 256: 2ad121402920a53918c6a31f30a9e93f9e0dc5cac94545edd78454f96bd36996

The numbers at a glance

Arena ratingfrom votes by users · higher is better

n/a

PixArt-Sigma-XL-2-1024-MS

not in the arena

n/a

HunyuanImage 3.0

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better

PixArt-Sigma-XL-2-1024-MS

Rank 99 of 108

HunyuanImage 3.0

Rank 89 of 108

Price per imagedollars through the API · lower is better

n/a

PixArt-Sigma-XL-2-1024-MS

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.

PixArt-Sigma-XL-2-1024-MS

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	PixArt-Sigma-XL-2-1024-MS	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	13	27
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition	3	13
Product photographyMaterials, reflections, and fine details	16	34
Character and detailAnatomy, clothing, and spatial consistency	27	24
InfographicNumbers, labels, and visual organization	5	36
API price
Per imagelower is better	n/a	n/a
100 images	n/a	n/a
Profile
Released	Apr 9, 2024	Sep 28, 2025
Weightsopen means: runs locally	open	open
Access	local	local
Parameters	0.6 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: PixArt-Sigma-XL-2-1024-MS 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 PixArt-Sigma-XL-2-1024-MS and HunyuanImage 3.0 from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare PixArt-Sigma-XL-2-1024-MS and HunyuanImage 3.0 with the models people compare them with most often.

Compare another matchup

First image modelSecond image model