Stable Diffusion 3.5 Largevs.Kandinsky-5.0-T2I-Lite-SFT

Which one is better?

In our own test both are on par. Kandinsky-5.0-T2I-Lite-SFT has no arena rating yet. Neither has an API price.

Test images

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

Stable Diffusion 3.5 Large: Character and detail, first image from the fixed test run — Stable Diffusion 3.5 Large23 out of 100

Kandinsky-5.0-T2I-Lite-SFT: Character and detail, first image from the fixed test run — Kandinsky-5.0-T2I-Lite-SFT65 out of 100

Result for “Character and detail”:Kandinsky-5.0-T2I-Lite-SFT ahead

Prompt given to both modelsCreate a square cinematic full-body portrait of a fictional bicycle courier waiting under a transparent umbrella at a rainy tram stop in Hamburg at blue hour. She wears a mustard raincoat, navy trousers, red sneakers, a silver helmet, and carries a teal messenger bag. Keep both hands visible, render the bicycle correctly, use realistic wet-street reflections, and include no readable brand names.

What the rating looks at: Anatomy, clothing, and spatial consistency.

“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

Stable Diffusion 3.5 Large: fal, requested 1024x1024, n=1, seed=7087247023770270000, estimated cost $0.068157, generated Aug 14, 2026, 2:37 PM. Source

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

Stable Diffusion 3.5 Large image SHA 256: e748006fe56e442ad347e3f994f18b86ee1739f20dcee063aa7a03ee710c5d54

Kandinsky-5.0-T2I-Lite-SFT image SHA 256: 8d193e702ad1aa0fccdb45f5d1474d60e21d455fa58d13470edd2796af0f2aed

Infographic

Stable Diffusion 3.5 Large: fal, requested 1024x1024, n=1, seed=12166653711354006000, estimated cost $0.068157, generated Aug 14, 2026, 2:37 PM. Source

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

Stable Diffusion 3.5 Large image SHA 256: 4eb0fd741ef42e6d3fee4b1280cdfb098a0a04009b139051963895892e349f1e

Kandinsky-5.0-T2I-Lite-SFT image SHA 256: e102016732db9e8e463504a83ac0770ff8a3c4c77918143d0f9dae8498c79101

Product photography

Stable Diffusion 3.5 Large: fal, requested 1024x1024, n=1, seed=13346308312015933000, estimated cost $0.068157, generated Aug 14, 2026, 2:37 PM. Source

Kandinsky-5.0-T2I-Lite-SFT: manual, local PyTorch 2.13.0, Diffusers 0.36.0; kandinskylab/Kandinsky-5.0-T2I-Lite-sft-Diffusers@c0158b5b20858d1fc02088db7ed4d805a296498b; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.5, seed=4636200764941689006; mps for text encoder and denoising; cpu for native VAE decode; bfloat16 for text encoders; float32 for transformer and CPU VAE; native tokenizer truncation: CLIP 80 to 77 tokens; archived input unchanged, generated Nov 2, 2026, 8:02 PM. Source

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

Stable Diffusion 3.5 Large image SHA 256: ea9a7401dee8859ec2cb96af9f7ca29a177ef2202f52f629d8ce9f5839bd4061

Kandinsky-5.0-T2I-Lite-SFT image SHA 256: 73b31dc5a48059f7712c6a125ac4855f2401133101172c4bfbaa11b76c5ceee1

Typography and layout

Stable Diffusion 3.5 Large: fal, requested 1024x1024, n=1, seed=16632333557910893000, estimated cost $0.068157, generated Aug 14, 2026, 2:37 PM. Source

Kandinsky-5.0-T2I-Lite-SFT: manual, local PyTorch 2.13.0, Diffusers 0.36.0; kandinskylab/Kandinsky-5.0-T2I-Lite-sft-Diffusers@c0158b5b20858d1fc02088db7ed4d805a296498b; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.5, seed=378697477899808767; mps for text encoder and denoising; cpu for native VAE decode; bfloat16 for text encoders; float32 for transformer and CPU VAE; three local execution attempts: first aborted on an MPS dtype error; second returned an unsaved latent but the runner read the wrong output field; accepted first saved image after correcting that field with unchanged prompt, seed and FP32 sampling settings; lost latent content remains unknown, generated Nov 2, 2026, 7:52 PM. Source

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

Stable Diffusion 3.5 Large image SHA 256: e2994b1883a245b1c8c447957d121ea89d93aafff149029aa4e8507dbb769f9a

Kandinsky-5.0-T2I-Lite-SFT image SHA 256: a84280dd219692f7b20722a37491d6788d1ecdc594f816fa1203981642501ab2

The numbers at a glance

Arena ratingfrom votes by users · higher is better

938

Stable Diffusion 3.5 Large

Rank 21 of 21

n/a

Kandinsky-5.0-T2I-Lite-SFT

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better

Stable Diffusion 3.5 Large

Rank 87 of 108

Kandinsky-5.0-T2I-Lite-SFT

Rank 79 of 108

Price per imagedollars through the API · lower is better

n/a

Stable Diffusion 3.5 Large

no API price

n/a

Kandinsky-5.0-T2I-Lite-SFT

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.

Stable Diffusion 3.5 Large

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

Kandinsky-5.0-T2I-Lite-SFT

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 · 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	Stable Diffusion 3.5 Large	Kandinsky-5.0-T2I-Lite-SFT
Ratings
Arena ratingfrom votes by users, higher is better	938	n/a
Arena rank	Rank 21 of 21	n/a
Arena votesmore votes, a safer rating	23,396	n/a
GraduallyBenchour own rating, 0 to 100	27	34
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition	31	21
Product photographyMaterials, reflections, and fine details	17	15
Character and detailAnatomy, clothing, and spatial consistency	23	65
InfographicNumbers, labels, and visual organization	39	33
API price
Per imagelower is better	n/a	n/a
100 images	n/a	n/a
Profile
Released	Oct 22, 2024	n/a
Weightsopen means: runs locally	open	open
Access	API, local	local
Parameters	8.1 billion	6 billion
Largest output	1 Megapixel	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: Stable Diffusion 3.5 Large model source (verified Aug 15, 2026) · Kandinsky-5.0-T2I-Lite-SFT 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 Stable Diffusion 3.5 Large and Kandinsky-5.0-T2I-Lite-SFT from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare Stable Diffusion 3.5 Large and Kandinsky-5.0-T2I-Lite-SFT with the models people compare them with most often.

Compare another matchup

First image modelSecond image model

Test images

The numbers at a glance

Stable Diffusion 3.5 Large

Kandinsky-5.0-T2I-Lite-SFT

All data side by side

Arena ratings by category

No shared arena score in the stored snapshot

More comparisons

Compare another matchup

The most popular comparisons