MOSS-Image-SFTvs.Boogu-Image-0.1-Turbo

Which one is better?

In our own test, MOSS-Image-SFT 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-SFT: Product photography, first image from the fixed test run — MOSS-Image-SFT86 out of 100

Boogu-Image-0.1-Turbo: Product photography, first image from the fixed test run — Boogu-Image-0.1-Turbo6 out of 100

Result for “Product photography”:MOSS-Image-SFT ahead

Prompt given to both modelsCreate 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.

What the rating looks at: Materials, reflections, and fine details.

“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-SFT: fal, requested 1024x1024, n=1, seed=2105639597, prompt expansion disabled, estimated charge ceiling $0.055, Fal lists $0.05 per MP, estimated cost $0.055000, generated Nov 1, 2026, 8:45 PM. Source

Boogu-Image-0.1-Turbo: manual, local PyTorch 2.11.0, Diffusers 0.38.0; Boogu/Boogu-Image-0.1-Turbo@47cd26d3c211b030c64ffe60d63b2e17a7c519c2; native 1024x1024, 1:1; one request, first output; steps=4, guidance=1.0, seed=2111621457284287488; mps; native PyTorch fallback is enabled by PYTORCH_ENABLE_MPS_FALLBACK=1; actual CPU operation use is not measured; bfloat16; upstream boogu-project/Boogu-Image@9b327526b1053e1f27f1bd0ae36b20ce2c39cb9e; reviewed MPS device allowlist and empty-reference self-attention only on original CPU SDPA, unchanged arguments and dtype; nonempty attention remains on MPS; no CUDA numeric equivalence claimed; native VAE input saved after scaling and shift, decoded float output finite before native same-size interpolation; native instruction template without rewriting; original DMD student inference, conditioning sigma 0.001, generated Nov 2, 2026, 11:24 PM. Source

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

MOSS-Image-SFT image SHA 256: 772013206361c55e807d5004620682a603061ffbabb0cb28e24993a18c0a4ad5

Boogu-Image-0.1-Turbo image SHA 256: 0c90d3a3e12870784d8b576ac3ace689b6cb627a4222f7cfc4a0d3793b9e4a37

Infographic

MOSS-Image-SFT: fal, requested 1024x1024, n=1, seed=454165776, prompt expansion disabled, estimated charge ceiling $0.055, Fal lists $0.05 per MP, estimated cost $0.055000, generated Nov 1, 2026, 8:45 PM. Source

Boogu-Image-0.1-Turbo: manual, local PyTorch 2.11.0, Diffusers 0.38.0; Boogu/Boogu-Image-0.1-Turbo@47cd26d3c211b030c64ffe60d63b2e17a7c519c2; native 1024x1024, 1:1; one request, first output; steps=4, guidance=1.0, seed=8524023330556932596; mps; native PyTorch fallback is enabled by PYTORCH_ENABLE_MPS_FALLBACK=1; actual CPU operation use is not measured; bfloat16; upstream boogu-project/Boogu-Image@9b327526b1053e1f27f1bd0ae36b20ce2c39cb9e; reviewed MPS device allowlist and empty-reference self-attention only on original CPU SDPA, unchanged arguments and dtype; nonempty attention remains on MPS; no CUDA numeric equivalence claimed; native VAE input saved after scaling and shift, decoded float output finite before native same-size interpolation; native instruction template without rewriting; original DMD student inference, conditioning sigma 0.001, generated Nov 2, 2026, 11:24 PM. Source

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

MOSS-Image-SFT image SHA 256: f46b17df7becc603d27762b2d235128ffc30f5722f51e8d50cd6df4723409dd9

Boogu-Image-0.1-Turbo image SHA 256: 096ae82e5a28ef1095f159019c41983c84a0c670a76e752fb3a1f9600fa7c45a

Product photography

MOSS-Image-SFT: fal, requested 1024x1024, n=1, seed=1064044164, prompt expansion disabled, estimated charge ceiling $0.055, Fal lists $0.05 per MP, estimated cost $0.055000, generated Nov 1, 2026, 8:45 PM. Source

Boogu-Image-0.1-Turbo: manual, local PyTorch 2.11.0, Diffusers 0.38.0; Boogu/Boogu-Image-0.1-Turbo@47cd26d3c211b030c64ffe60d63b2e17a7c519c2; native 1024x1024, 1:1; one request, first output; steps=4, guidance=1.0, seed=9219148100655939974; mps; native PyTorch fallback is enabled by PYTORCH_ENABLE_MPS_FALLBACK=1; actual CPU operation use is not measured; bfloat16; upstream boogu-project/Boogu-Image@9b327526b1053e1f27f1bd0ae36b20ce2c39cb9e; reviewed MPS device allowlist and empty-reference self-attention only on original CPU SDPA, unchanged arguments and dtype; nonempty attention remains on MPS; no CUDA numeric equivalence claimed; native VAE input saved after scaling and shift, decoded float output finite before native same-size interpolation; native instruction template without rewriting; original DMD student inference, conditioning sigma 0.001, generated Nov 2, 2026, 11:23 PM. Source

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

MOSS-Image-SFT image SHA 256: ae1015034efbde73cece799521a3a2e98ad82a8362eaf97f5bc1266e13386d57

Boogu-Image-0.1-Turbo image SHA 256: 50719aadf378d21848c83dbdb510a8134c80be05cbb2b718182a95e9cbb28164

Typography and layout

MOSS-Image-SFT: fal, requested 1024x1024, n=1, seed=1635014357, prompt expansion disabled, estimated charge ceiling $0.055, Fal lists $0.05 per MP, estimated cost $0.055000, generated Nov 1, 2026, 8:44 PM. Source

Boogu-Image-0.1-Turbo: manual, local PyTorch 2.11.0, Diffusers 0.38.0; Boogu/Boogu-Image-0.1-Turbo@47cd26d3c211b030c64ffe60d63b2e17a7c519c2; native 1024x1024, 1:1; one request, first output; steps=4, guidance=1.0, seed=4970356096182813142; mps; native PyTorch fallback is enabled by PYTORCH_ENABLE_MPS_FALLBACK=1; actual CPU operation use is not measured; bfloat16; upstream boogu-project/Boogu-Image@9b327526b1053e1f27f1bd0ae36b20ce2c39cb9e; reviewed MPS device allowlist and empty-reference self-attention only on original CPU SDPA, unchanged arguments and dtype; nonempty attention remains on MPS; no CUDA numeric equivalence claimed; native VAE input saved after scaling and shift, decoded float output finite before native same-size interpolation; native instruction template without rewriting; original DMD student inference, conditioning sigma 0.001; two local attempts: original MPSGraph SIGABRT before output; one identical-prompt, identical-seed technical repeat after reviewed backend-only correction, first output retained; exact failing original operator is not proven, generated Nov 2, 2026, 11:18 PM. Source

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

MOSS-Image-SFT image SHA 256: 61feaf0c1af07a1cc8dd874c51dc8e86b5bfc271042b8e25f2512cb5a03ec51d

Boogu-Image-0.1-Turbo image SHA 256: cec0e99f796e6d0e4dafc517afc93fd93041aecbd52cef74a421ea9c3a15ab69

The numbers at a glance

Arena ratingfrom votes by users · higher is better

n/a

MOSS-Image-SFT

not in the arena

n/a

Boogu-Image-0.1-Turbo

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better

MOSS-Image-SFT

Rank 43 of 108

Boogu-Image-0.1-Turbo

Rank 104 of 108

Price per imagedollars through the API · lower is better

n/a

MOSS-Image-SFT

no API price

n/a

Boogu-Image-0.1-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.

MOSS-Image-SFT

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

Boogu-Image-0.1-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.

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-SFT	Boogu-Image-0.1-Turbo
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	64	9
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition	22	18
Product photographyMaterials, reflections, and fine details	86	6
Character and detailAnatomy, clothing, and spatial consistency	74	5
InfographicNumbers, labels, and visual organization	75	8
API price
Per imagelower is better	n/a	n/a
100 images	n/a	n/a
Profile
Released	Oct 16, 2026	n/a
Weightsopen means: runs locally	open	open
Access	local	local
Parameters	5 billion	10 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-SFT model source (verified Nov 1, 2026) · Boogu-Image-0.1-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 MOSS-Image-SFT and Boogu-Image-0.1-Turbo from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare MOSS-Image-SFT and Boogu-Image-0.1-Turbo with the models people compare them with most often.

Compare another matchup

First image modelSecond image model