Skip to main content

Krea 2 Medium Turbovs.SenseNova U1.5 Lite

Which one is better?

In our own test both are on par. Neither is in the arena. SenseNova U1.5 Lite has no API price.

Test images

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

Krea 2 Medium Turbo: Typography and layout, first image from the fixed test run
Krea 2 Medium Turbo71 out of 100
SenseNova U1.5 Lite: Typography and layout, first image from the fixed test run
SenseNova U1.5 Lite93 out of 100

Result for “Typography and layout”:SenseNova U1.5 Lite 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

Krea 2 Medium Turbo: openrouter, aspect ratio 1:1, resolution 1K, one request, first returned image, seed=provider random, actual cost $0.015000, generated Aug 14, 2026, 1:35 PM. Source

SenseNova U1.5 Lite: manual, local stable-diffusion.cpp@3be23788fc0c9f35a2c24b1211caae11d4a54dbc, ggml@d49c49e1dd361ba3c0a77c90b5b1404b83969aa8; sensenova/SenseNova-U1.5-8B-MoT@9feeeab8a2792514d109cd34589342a2cc1d4ab2; native 2048x2048, 1:1; one request, first output; steps=50, guidance=4.0, seed=8087247682796819534; Apple M5 Max Metal MTL0 requested through the native backend option; automatic placement remains enabled, and actual placement is recorded in the CLI log; Original BF16 checkpoint. The exact two SenseNova vision convolution weights use a targeted lossless BF16-to-F32 rule; other eligible weights use the native unquantized f16 option. This BF16-to-F16 conversion rounds values, as quantified by the linked complete-checkpoint GGML preflight; excluded weights remain BF16 if used. Operation precision follows the backend.; Euler, flow shift 3; CPU Philox RNG (CLI cuda RNG option, no CUDA hardware); original BF16 weights without requested quantization; observational latent and pre-clipping finite checks; native prompt template without rewriting; empty unconditional prompt, source 2048x2048, stored 1024x1024 (Lanczos3 downscale), generated Nov 2, 2026, 10:17 PM. Source

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

Krea 2 Medium Turbo image SHA 256: a51c5eeb5ba6b3ab8be3cb1d9c36fc196d71f4034ff7e57c695af1fe2f4ab078

SenseNova U1.5 Lite image SHA 256: c092b333f479c5b28ad33958d0e45f0fb0989970af02690079cf1398742c55b8

Infographic

Krea 2 Medium Turbo: openrouter, aspect ratio 1:1, resolution 1K, one request, first returned image, seed=provider random, actual cost $0.015000, generated Aug 14, 2026, 1:35 PM. Source

SenseNova U1.5 Lite: manual, local stable-diffusion.cpp@3be23788fc0c9f35a2c24b1211caae11d4a54dbc, ggml@d49c49e1dd361ba3c0a77c90b5b1404b83969aa8; sensenova/SenseNova-U1.5-8B-MoT@9feeeab8a2792514d109cd34589342a2cc1d4ab2; native 2048x2048, 1:1; one request, first output; steps=50, guidance=4.0, seed=2640628636372692680; Apple M5 Max Metal MTL0 requested through the native backend option; automatic placement remains enabled, and actual placement is recorded in the CLI log; Original BF16 checkpoint. The exact two SenseNova vision convolution weights use a targeted lossless BF16-to-F32 rule; other eligible weights use the native unquantized f16 option. This BF16-to-F16 conversion rounds values, as quantified by the linked complete-checkpoint GGML preflight; excluded weights remain BF16 if used. Operation precision follows the backend.; Euler, flow shift 3; CPU Philox RNG (CLI cuda RNG option, no CUDA hardware); original BF16 weights without requested quantization; observational latent and pre-clipping finite checks; native prompt template without rewriting; empty unconditional prompt, source 2048x2048, stored 1024x1024 (Lanczos3 downscale), generated Nov 2, 2026, 10:25 PM. Source

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

Krea 2 Medium Turbo image SHA 256: 10398e0fce4a5b07b8a38a010eec70ac5457d0209ef67979feebb111d315fe16

SenseNova U1.5 Lite image SHA 256: dcf8206386da05d11c35a6950309e61f08a504d440366226eac316328718f5ae

Product photography

Krea 2 Medium Turbo: openrouter, aspect ratio 1:1, resolution 1K, one request, first returned image, seed=provider random, actual cost $0.015000, generated Aug 14, 2026, 1:35 PM. Source

SenseNova U1.5 Lite: manual, local stable-diffusion.cpp@3be23788fc0c9f35a2c24b1211caae11d4a54dbc, ggml@d49c49e1dd361ba3c0a77c90b5b1404b83969aa8; sensenova/SenseNova-U1.5-8B-MoT@9feeeab8a2792514d109cd34589342a2cc1d4ab2; native 2048x2048, 1:1; one request, first output; steps=50, guidance=4.0, seed=1950293105094404455; Apple M5 Max Metal MTL0 requested through the native backend option; automatic placement remains enabled, and actual placement is recorded in the CLI log; Original BF16 checkpoint. The exact two SenseNova vision convolution weights use a targeted lossless BF16-to-F32 rule; other eligible weights use the native unquantized f16 option. This BF16-to-F16 conversion rounds values, as quantified by the linked complete-checkpoint GGML preflight; excluded weights remain BF16 if used. Operation precision follows the backend.; Euler, flow shift 3; CPU Philox RNG (CLI cuda RNG option, no CUDA hardware); original BF16 weights without requested quantization; observational latent and pre-clipping finite checks; native prompt template without rewriting; empty unconditional prompt, source 2048x2048, stored 1024x1024 (Lanczos3 downscale), generated Nov 2, 2026, 10:11 PM. Source

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

Krea 2 Medium Turbo image SHA 256: e1511de7ba316289e69ae79e51b4f75c80ee1468611f4cebcfc9d21658b19cb8

SenseNova U1.5 Lite image SHA 256: d425ed5236325ee28633f83cb00f3fd39181f7dd6df0e5408d8b1c07c6796107

Typography and layout

Krea 2 Medium Turbo: openrouter, aspect ratio 1:1, resolution 1K, one request, first returned image, seed=provider random, actual cost $0.015000, generated Aug 14, 2026, 2:05 PM. Source

SenseNova U1.5 Lite: manual, local stable-diffusion.cpp@3be23788fc0c9f35a2c24b1211caae11d4a54dbc, ggml@d49c49e1dd361ba3c0a77c90b5b1404b83969aa8; sensenova/SenseNova-U1.5-8B-MoT@9feeeab8a2792514d109cd34589342a2cc1d4ab2; native 2048x2048, 1:1; one request, first output; steps=50, guidance=4.0, seed=8328647217678382438; Apple M5 Max Metal MTL0 requested through the native backend option; automatic placement remains enabled, and actual placement is recorded in the CLI log; Original BF16 checkpoint. The exact two SenseNova vision convolution weights use a targeted lossless BF16-to-F32 rule; other eligible weights use the native unquantized f16 option. This BF16-to-F16 conversion rounds values, as quantified by the linked complete-checkpoint GGML preflight; excluded weights remain BF16 if used. Operation precision follows the backend.; Euler, flow shift 3; CPU Philox RNG (CLI cuda RNG option, no CUDA hardware); original BF16 weights without requested quantization; observational latent and pre-clipping finite checks; native prompt template without rewriting; empty unconditional prompt; three local execution attempts: original Metal BF16 attempt aborted on a missing matrix kernel before output; same-seed CPU retry deliberately interrupted before its first sampling step without image or latent; first image retained after the reviewed loader compatibility correction and disclosed F16 runtime conversion with unchanged prompt and seed, source 2048x2048, stored 1024x1024 (Lanczos3 downscale), generated Nov 2, 2026, 10:04 PM. Source

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

Krea 2 Medium Turbo image SHA 256: 50b1b1bb7196a612240ab6088330985474c9f60ccaf640e4aa8debb0aa20e1c8

SenseNova U1.5 Lite image SHA 256: df5c50fc54f03b972796f424999bd5991ff758a6b7abc00c212864aca3b124bc

The numbers at a glance

Arena ratingfrom votes by users · higher is better
n/a

Krea 2 Medium Turbo

not in the arena

n/a

SenseNova U1.5 Lite

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better
65

Krea 2 Medium Turbo

Rank 39 of 108

73

SenseNova U1.5 Lite

Rank 14 of 108

Price per imagedollars through the API · lower is better
$0.015

Krea 2 Medium Turbo

cheap

n/a

SenseNova U1.5 Lite

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.

Krea 2 Medium Turbo

Subscription
Free to use at Krea (Krea Free, with limited credits), more from ≈ €9.54 a month (Krea Basic, with limited credits). Through other providers from ≈ €12.72 a month (Leonardo.Ai Essential). 15 plans in Germany.
Via API
$0.015 per image, 100 images $1.50.

SenseNova U1.5 Lite

Subscription
n/a No plan in Germany names exactly this model.
Via API
No API price known. Access: API, Web app, 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 · API prices from Nov 6, 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.

KennzahlKrea 2 Medium TurboSenseNova U1.5 Lite
Ratings
Arena ratingfrom votes by users, higher is bettern/an/a
Arena rankn/an/a
Arena votesmore votes, a safer ratingn/an/a
GraduallyBenchour own rating, 0 to 1006573
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition7193
Product photographyMaterials, reflections, and fine details8184
Character and detailAnatomy, clothing, and spatial consistency3035
InfographicNumbers, labels, and visual organization7980
API price
Per imagelower is better$0.015n/a
100 images$1.50n/a
Profile
ReleasedJune 2026n/a
Weightsopen means: runs locallyproprietaryopen
AccessAPI, Web appAPI, Web app, local
Parametersnot publishednot published
Largest output1K4K
Plans in Germanynaming exactly this model15n/a

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

Sources: Krea 2 Medium Turbo model source (verified Aug 15, 2026) · SenseNova U1.5 Lite model source (verified Nov 2, 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 Krea 2 Medium Turbo and SenseNova U1.5 Lite from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare Krea 2 Medium Turbo and SenseNova U1.5 Lite with the models people compare them with most often.

Compare another matchup