Skip to main content

SenseTimeImage model

SenseNova U1.5 Lite

SenseNova U1.5 Lite is rank 14 of 108 in our GraduallyBench test.

Arena ratingnot rated

n/aUser voteshigher is better

GraduallyBenchRank 14 of 108

73our own rating0 to 100

Priceno price

n/aper imagelower is better

PopularityRank 670 of 875

0out of 100first value

Summary

SenseNova U1.5 Lite is an image model from SenseTime. The API can automatically extend prompts through prompt_extend.

Image samples from our tests

These images come from documented single requests with an archived prompt. They complement measurements but do not replace a controlled laboratory test.

Character and detailOpen Character and detail at original size

Character and detail

This sample tests Anatomy, clothing, and spatial consistency.

Prompt and settings

Create 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.

Generated through manual. Archived settings were 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).

Open source
InfographicOpen Infographic at original size

Infographic

This sample tests Numbers, labels, and visual organization.

Prompt and settings

Create 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.

Generated through manual. Archived settings were 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).

Open source
Product photographyOpen Product photography at original size

Product photography

This sample tests Materials, reflections, and fine details.

Prompt and settings

Create 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.

Generated through manual. Archived settings were 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).

Open source
Typography and layoutOpen Typography and layout at original size

Typography and layout

This sample tests Legible text, hierarchy, and composition.

Prompt and settings

Create 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.

Generated through manual. Archived settings were 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).

Open source

All benchmark results

Every published result with its source, test setup and comparison models.

Show all 4 published benchmark results

Published benchmarks

Gradually sample test: Typography and layout

The 0-100 value normalizes the mean rank from model-blind comparison rounds. One archived first output contributes per model. This remains an n = 1 sample.

Nano Banana 299.7 rank points
FLUX.2 Flex98.1 rank points
FLUX 3 Image96 rank points
Recraft V4.1 Utility95.6 rank points
Wan 2.7 Image94.1 rank points
SenseNova U1.5 Lite93.1 rank points
Kling Image 3.0 Omni92.8 rank points
Nano Banana Pro91.9 rank points
FLUX.2 Max91.6 rank points
Recraft V4.1 Flash88.5 rank points
Krea 2 Large86.3 rank points
Firefly Image Model 585.7 rank points
Muse Image85.7 rank points
FLUX 3 Image Klein84.4 rank points
Seedream 4.583.8 rank points
MAI-Image-2.582.2 rank points
Nano Banana 2 Lite82.2 rank points
Qwen Image 2.0 Pro81 rank points
Wan 2.7 Image Pro81 rank points
Krea 377.6 rank points
Seedream 5.0 Flash76.6 rank points
HiDream-O1-Image-1.575.7 rank points
Seedream 5.0 Lite75.7 rank points
Luma Uni-1.1 Max75.1 rank points
GPT Image 274.5 rank points
Grok Imagine Image 1.073.8 rank points
Nano Banana 2.172.3 rank points
Krea 2 Medium Turbo71.3 rank points
Grok Imagine Image 2.070.7 rank points
MAI-Image-2.5 Flash70.1 rank points
Recraft V4.169.5 rank points
FLUX.2 Pro68.9 rank points
Bria FIBO 268.5 rank points
GPT Image 1 Mini67.9 rank points
Seedream 5.0 Pro67.9 rank points
Ideogram 4.067.3 rank points
Luma Uni-1.167 rank points
Seedream 4.064.8 rank points
GPT Image 2.5 Flare63.9 rank points
LLaDA-Image62 rank points
MAI-Image-2.662 rank points
GPT Image 1.559.5 rank points
FLUX.2 Klein 9B Preview55.8 rank points
Reve 2.153.9 rank points
GPT Image 2.5 Sunburst53.6 rank points
MAI-Image-2.6 Flash53.6 rank points
Qwen-Image-2.153.6 rank points
Kling Image 3.051.4 rank points
Runway Gen-4 Image51.4 rank points
MOSS-Image-Base50.8 rank points
Qwen Image 2.1 Turbo50.8 rank points
GLM-Image50.2 rank points
FLUX.2 Dev49.8 rank points
FLUX.1 Kontext [max]45.5 rank points
Hy Image 3.5 Preview45.2 rank points
FLUX1.1 [pro]42.1 rank points
LongCat-Image41.4 rank points
Wan 2.6 Text to Image41.1 rank points
Z-Image Turbo39.6 rank points
Nucleus-Image38.3 rank points
Qwen Image 2.1 Pro38.3 rank points
MiniMax Image-0137.4 rank points
Chroma1-HD36.4 rank points
Bria FIBO Lite35.5 rank points
Qwen Image 3.0 Pro35.5 rank points
Z-Image Base32.7 rank points
FLUX.2 Klein 9B32.4 rank points
Qwen Image 3.032.4 rank points
FLUX.2 Klein 4B31.8 rank points
OmniGen229.6 rank points
Ideogram 4.528.3 rank points
Vidu Q2 Image26.8 rank points
FLUX1.1 [pro] Ultra21.8 rank points
MOSS-Image-SFT21.5 rank points
Adobe Firefly Image 420.6 rank points
Sana-Sprint 1.6B 1024px20.6 rank points
FLUX.1 Kontext [pro]19 rank points
Lumina-Image 2.018.4 rank points
Boogu-Image-0.1-Turbo18.1 rank points
HiDream-O1-Image18.1 rank points
Chroma1-Flash16.2 rank points
FLUX.1 [dev]16.2 rank points
Sana-Sprint 0.6B 1024px15.9 rank points
Midjourney V8.114.3 rank points
HunyuanImage 3.012.5 rank points
AuraFlow v0.311.2 rank points
BAGEL-7B-MoT11.2 rank points
SANA-1.5 4.8B 1024px8.1 rank points
LLaDA-Image-Turbo7.8 rank points
Adobe Firefly Image 36.9 rank points
SDXL Turbo3.4 rank points
SANA-1.5 1.6B 1024px3.1 rank points

higher is betterAs of November 3, 2026Scale: 0 rank points - 100 rank points

Source: Gradually-Bildtest
gradually.ai
Methodology and context
Metric
rank-score-100
Scale
0-100
Participants
108
Sample
1
Methodology and context
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-Bildtest

Gradually sample test: Product photography

The 0-100 value normalizes the mean rank from model-blind comparison rounds. One archived first output contributes per model. This remains an n = 1 sample.

MAI-Image-2.6 Flash90.3 rank points
Krea 2 Large90 rank points
Qwen Image 2.1 Turbo89.7 rank points
Qwen-Image-2.189.7 rank points
GPT Image 289.4 rank points
Seedream 5.0 Lite88.8 rank points
MAI-Image-2.688.2 rank points
BAGEL-7B-MoT87.9 rank points
Wan 2.7 Image87.9 rank points
Nano Banana 2 Lite87.2 rank points
Qwen Image 3.087.2 rank points
FLUX.2 Klein 9B86.6 rank points
GPT Image 2.5 Sunburst86.6 rank points
Wan 2.7 Image Pro86.6 rank points
MOSS-Image-SFT85.7 rank points
Bria FIBO 283.8 rank points
SenseNova U1.5 Lite83.5 rank points
Qwen Image 2.1 Pro83.2 rank points
Grok Imagine Image 1.081.3 rank points
FLUX.2 Klein 4B80.7 rank points
Krea 2 Medium Turbo80.7 rank points
Muse Image80.7 rank points
Seedream 5.0 Flash76.9 rank points
Seedream 5.0 Pro76.9 rank points
FLUX 3 Image Klein74.8 rank points
Reve 2.173.8 rank points
GPT Image 1 Mini73.5 rank points
FLUX.2 Klein 9B Preview72.6 rank points
Qwen Image 3.0 Pro71.3 rank points
Luma Uni-1.1 Max71 rank points
Recraft V4.171 rank points
MAI-Image-2.570.1 rank points
Nano Banana Pro70.1 rank points
Nano Banana 2.169.8 rank points
GPT Image 2.5 Flare69.2 rank points
Ideogram 4.568.9 rank points
Grok Imagine Image 2.067.9 rank points
FLUX.2 Flex62.6 rank points
Krea 362.6 rank points
FLUX 3 Image62 rank points
Nucleus-Image62 rank points
Kling Image 3.060.4 rank points
Luma Uni-1.159.5 rank points
FLUX.2 Max58.6 rank points
Qwen Image 2.0 Pro58.3 rank points
FLUX.2 Pro57.9 rank points
MAI-Image-2.5 Flash57.6 rank points
Wan 2.6 Text to Image53.9 rank points
GPT Image 1.552.3 rank points
HiDream-O1-Image-1.550.8 rank points
Midjourney V8.149.5 rank points
FLUX.1 [dev]46.4 rank points
FLUX.2 Dev46.4 rank points
Firefly Image Model 545.8 rank points
Nano Banana 244.9 rank points
Seedream 4.044.2 rank points
Seedream 4.544.2 rank points
Kling Image 3.0 Omni39.9 rank points
MOSS-Image-Base36.1 rank points
Recraft V4.1 Flash36.1 rank points
FLUX.1 Kontext [pro]35.8 rank points
Recraft V4.1 Utility Pro35.8 rank points
Bria FIBO Lite35.2 rank points
LLaDA-Image35.2 rank points
HunyuanImage 3.033.6 rank points
Runway Gen-4 Image32.7 rank points
Recraft V4.1 Utility32.4 rank points
Vidu Q2 Image32.4 rank points
HiDream-O1-Image31.1 rank points
MiniMax Image-0129.9 rank points
OmniGen229.9 rank points
Z-Image Turbo29.9 rank points
LongCat-Image29.3 rank points
FLUX.1 Kontext [max]28.7 rank points
FLUX1.1 [pro] Ultra28.3 rank points
LLaDA-Image-Turbo26.8 rank points
Sana-Sprint 1.6B 1024px26.8 rank points
FLUX1.1 [pro]25.5 rank points
Ideogram 4.022.1 rank points
Z-Image Base21.5 rank points
Chroma1-HD15.9 rank points
Lumina-Image 2.015.9 rank points
Chroma1-Flash15.6 rank points
Hy Image 3.5 Preview15.3 rank points
AuraFlow v0.315 rank points
Adobe Firefly Image 314.6 rank points
Adobe Firefly Image 413.7 rank points
Stable Diffusion XL 1.012.8 rank points
GLM-Image9 rank points
Boogu-Image-0.1-Base7.8 rank points
Boogu-Image-0.1-Turbo5.6 rank points
SDXL Turbo2.8 rank points
SANA-1.5 4.8B 1024px1.2 rank points
Stable Diffusion Turbo1.2 rank points
SANA-1.5 1.6B 1024px0.3 rank points

higher is betterAs of November 3, 2026Scale: 0 rank points - 100 rank points

Source: Gradually-Bildtest
gradually.ai
Methodology and context
Metric
rank-score-100
Scale
0-100
Participants
108
Sample
1
Methodology and context
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-Bildtest

Gradually sample test: Character and detail

The 0-100 value normalizes the mean rank from model-blind comparison rounds. One archived first output contributes per model. This remains an n = 1 sample.

GPT Image 1.588.8 rank points
Qwen Image 2.0 Pro88.5 rank points
FLUX.2 Pro86.9 rank points
Krea 384.7 rank points
FLUX.2 Dev81.6 rank points
Kling Image 3.0 Omni81.6 rank points
Reve 2.180.1 rank points
Wan 2.6 Text to Image78.5 rank points
Wan 2.7 Image Pro78.2 rank points
Seedream 5.0 Flash75.4 rank points
FLUX.2 Flex75.1 rank points
Grok Imagine Image 2.075.1 rank points
MOSS-Image-SFT74.1 rank points
Qwen Image 3.0 Pro74.1 rank points
Seedream 4.072.9 rank points
Firefly Image Model 572.3 rank points
Z-Image Base71 rank points
Z-Image Turbo71 rank points
Bria FIBO Lite70.7 rank points
FLUX.2 Max70.4 rank points
FLUX 3 Image69.5 rank points
MAI-Image-2.5 Flash67.9 rank points
Ideogram 4.567.6 rank points
FLUX 3 Image Klein66.4 rank points
Nucleus-Image66.4 rank points
Recraft V4.1 Utility66.4 rank points
Qwen Image 2.1 Pro64.8 rank points
Recraft V4.1 Utility Pro64.8 rank points
GPT Image 2.5 Sunburst64.2 rank points
GPT Image 2.5 Flare63.2 rank points
Wan 2.7 Image63.2 rank points
MAI-Image-2.662.9 rank points
LLaDA-Image62.6 rank points
Qwen Image 3.061.7 rank points
FLUX.2 Klein 4B61.1 rank points
Muse Image61.1 rank points
Grok Imagine Image 1.060.7 rank points
Luma Uni-1.1 Max60.4 rank points
GPT Image 260.1 rank points
Seedream 5.0 Pro60.1 rank points
Kling Image 3.059.2 rank points
Ideogram 4.058.9 rank points
HiDream-O1-Image-1.558.3 rank points
Nano Banana 2.156.7 rank points
MAI-Image-2.554.8 rank points
Krea 2 Large54.2 rank points
FLUX.1 [dev]52.6 rank points
LLaDA-Image-Turbo52.6 rank points
Adobe Firefly Image 452.3 rank points
MAI-Image-2.6 Flash52 rank points
Midjourney V8.151.4 rank points
GLM-Image51.1 rank points
Chroma1-Flash50.8 rank points
FLUX.1 Kontext [pro]50.2 rank points
Hy Image 3.5 Preview49.5 rank points
Luma Uni-1.148.6 rank points
Seedream 5.0 Lite48.6 rank points
Bria FIBO 248.3 rank points
Recraft V4.148 rank points
Qwen Image 2.1 Turbo47.7 rank points
Nano Banana 2 Lite47.4 rank points
MOSS-Image-Base45.5 rank points
FLUX.2 Klein 9B Preview45.2 rank points
FLUX1.1 [pro] Ultra45.2 rank points
HiDream-O1-Image45.2 rank points
Recraft V4.1 Flash42.4 rank points
FLUX1.1 [pro]39.3 rank points
FLUX.2 Klein 9B38.9 rank points
Runway Gen-4 Image38.9 rank points
Nano Banana Pro37.1 rank points
Seedream 4.537.1 rank points
Nano Banana 236.4 rank points
SenseNova U1.5 Lite34.9 rank points
Qwen-Image-2.133.6 rank points
MiniMax Image-0132.7 rank points
BAGEL-7B-MoT32.4 rank points
GPT Image 1 Mini30.8 rank points
Krea 2 Medium Turbo29.9 rank points
FLUX.1 Kontext [max]27.1 rank points
Adobe Firefly Image 326.8 rank points
Sana-Sprint 1.6B 1024px26.2 rank points
Vidu Q2 Image26.2 rank points
HunyuanImage 3.024.3 rank points
LongCat-Image24.3 rank points
Lumina-Image 2.023.4 rank points
Chroma1-HD16.5 rank points
AuraFlow v0.316.2 rank points
OmniGen27.8 rank points
Boogu-Image-0.1-Base7.5 rank points
Stable Diffusion Turbo4.7 rank points
SDXL Turbo2.8 rank points
SANA-1.5 4.8B 1024px0.3 rank points

higher is betterAs of November 3, 2026Scale: 0 rank points - 100 rank points

Source: Gradually-Bildtest
gradually.ai
Methodology and context
Metric
rank-score-100
Scale
0-100
Participants
108
Sample
1
Methodology and context
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-Bildtest

Gradually sample test: Infographic

The 0-100 value normalizes the mean rank from model-blind comparison rounds. One archived first output contributes per model. This remains an n = 1 sample.

Seedream 5.0 Pro99.1 rank points
GPT Image 2.5 Sunburst97.2 rank points
GPT Image 2.5 Flare91 rank points
Luma Uni-1.189.4 rank points
Qwen Image 2.1 Turbo88.8 rank points
Qwen Image 3.088.8 rank points
Wan 2.7 Image Pro88.5 rank points
Muse Image86.3 rank points
Seedream 5.0 Flash86 rank points
Ideogram 4.585.4 rank points
Reve 2.184.7 rank points
Qwen-Image-2.182.9 rank points
MAI-Image-2.5 Flash81.3 rank points
SenseNova U1.5 Lite80.4 rank points
Wan 2.7 Image79.4 rank points
Krea 2 Medium Turbo78.8 rank points
Luma Uni-1.1 Max78.8 rank points
Bria FIBO 278.2 rank points
FLUX 3 Image Klein77.3 rank points
Wan 2.6 Text to Image77.3 rank points
Grok Imagine Image 1.076.9 rank points
Grok Imagine Image 2.076.6 rank points
HiDream-O1-Image-1.576.3 rank points
Krea 2 Large75.7 rank points
MOSS-Image-SFT75.4 rank points
FLUX.2 Klein 9B Preview74.8 rank points
Recraft V4.1 Utility Pro74.5 rank points
Krea 373.8 rank points
Recraft V4.1 Flash73.8 rank points
MAI-Image-2.6 Flash73.5 rank points
Recraft V4.173.2 rank points
Ideogram 4.072.6 rank points
FLUX.2 Dev72 rank points
GPT Image 1.571.7 rank points
Hy Image 3.5 Preview70.7 rank points
MOSS-Image-Base70.7 rank points
MAI-Image-2.569.2 rank points
FLUX 3 Image68.5 rank points
Seedream 4.067.3 rank points
GPT Image 266.4 rank points
Nano Banana 2 Lite65.7 rank points
Nano Banana Pro64.8 rank points
Nano Banana 2.164.2 rank points
Recraft V4.1 Utility63.9 rank points
Nano Banana 263.2 rank points
Qwen Image 2.1 Pro62 rank points
Qwen Image 2.0 Pro61.7 rank points
Qwen Image 3.0 Pro61.7 rank points
MAI-Image-2.660.7 rank points
FLUX.2 Flex57 rank points
Seedream 5.0 Lite55.8 rank points
FLUX.1 Kontext [max]53.9 rank points
FLUX.2 Max53.6 rank points
FLUX.1 Kontext [pro]52.6 rank points
Kling Image 3.051.4 rank points
FLUX.2 Pro44.6 rank points
HiDream-O1-Image42.7 rank points
FLUX.2 Klein 9B41.7 rank points
FLUX.1 [dev]39.9 rank points
Z-Image Turbo37.4 rank points
Vidu Q2 Image37.1 rank points
LongCat-Image36.8 rank points
HunyuanImage 3.036.1 rank points
LLaDA-Image-Turbo35.8 rank points
Midjourney V8.135.2 rank points
LLaDA-Image32.7 rank points
Runway Gen-4 Image32.4 rank points
GPT Image 1 Mini31.8 rank points
FLUX.2 Klein 4B30.2 rank points
GLM-Image29.9 rank points
Kling Image 3.0 Omni29.6 rank points
FLUX1.1 [pro] Ultra27.7 rank points
Seedream 4.526.5 rank points
MiniMax Image-0125.5 rank points
FLUX1.1 [pro]24 rank points
Z-Image Base24 rank points
Nucleus-Image23.4 rank points
Chroma1-HD22.4 rank points
Bria FIBO Lite19.6 rank points
OmniGen219 rank points
Boogu-Image-0.1-Base15.9 rank points
Lumina-Image 2.015.9 rank points
Firefly Image Model 515 rank points
Adobe Firefly Image 412.5 rank points
Chroma1-Flash12.1 rank points
Boogu-Image-0.1-Turbo8.4 rank points
SDXL Turbo8.4 rank points
SANA-1.5 1.6B 1024px7.2 rank points
AuraFlow v0.36.9 rank points
BAGEL-7B-MoT5.9 rank points
SANA-1.5 4.8B 1024px5.6 rank points
Adobe Firefly Image 35.3 rank points
Stable Diffusion Turbo4.7 rank points

higher is betterAs of November 3, 2026Scale: 0 rank points - 100 rank points

Source: Gradually-Bildtest
gradually.ai
Methodology and context
Metric
rank-score-100
Scale
0-100
Participants
108
Sample
1
Methodology and context
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Source: Gradually-Bildtest

Recorded measurement details

Gradually sample test: Typography and layout (Show measurement, test conditions, and source)
Source value
93.1
Score
93.1
Metric
rank-score-100
Unit
rank score
Direction
Higher is better
Source rank
6
Sample
1
Source type
gradually-sample-test
Evaluator
gemini-3.1-pro-preview
Retrieved at
2026-11-03
Methodology
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Mean rank
8.33
Individual ranks
6, 7, 12
Criterion scores
headline 1.33, subline 1.33, text-control 1.33, visual-brief 1.33, layout 1.33
Source Gradually-Bildtest
Gradually sample test: Product photography (Show measurement, test conditions, and source)
Source value
83.5
Score
83.5
Metric
rank-score-100
Unit
rank score
Direction
Higher is better
Source rank
18
Sample
1
Source type
gradually-sample-test
Evaluator
gemini-3.1-pro-preview
Retrieved at
2026-11-03
Methodology
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Mean rank
18.67
Individual ranks
25, 7, 24
Criterion scores
subject 2, time 2, watch-details 2, photography 2, exclusions 2
Source Gradually-Bildtest
Gradually sample test: Character and detail (Show measurement, test conditions, and source)
Source value
34.9
Score
34.9
Metric
rank-score-100
Unit
rank score
Direction
Higher is better
Source rank
81
Sample
1
Source type
gradually-sample-test
Evaluator
gemini-3.1-pro-preview
Retrieved at
2026-11-03
Methodology
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Mean rank
70.67
Individual ranks
81, 62, 69
Criterion scores
scene 1, wardrobe 1, anatomy 1, props 1, environment 1
Source Gradually-Bildtest
Gradually sample test: Infographic (Show measurement, test conditions, and source)
Source value
80.4
Score
80.4
Metric
rank-score-100
Unit
rank score
Direction
Higher is better
Source rank
15
Sample
1
Source type
gradually-sample-test
Evaluator
gemini-3.1-pro-preview
Retrieved at
2026-11-03
Methodology
One archived image per model, using the first output of the accepted test attempt. Technical repeats are disclosed for the affected image. All 108 images are ranked in three model-blind orderings. The 0-100 value normalizes the mean rank. This remains an n = 1 sample, and an automated judge may have visual preferences of its own.
Mean rank
22
Individual ranks
21, 17, 28
Criterion scores
title 1.67, sequence 1.67, labels 1.33, diagram 1.67, clarity 1
Source Gradually-Bildtest

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

The 100 most popular models of this category, 0 to 100, higher is more popular

20 of 101 models

As of November 2026 · Source: gradually.ai popularity index

Over time

First value from November 5, 2026. The timeline appears from the second week.

SenseNova U1.5 Lite compared

Compare with another model

The most popular comparisons

  1. 1GPT Image 2.5 Sunburst vs. SenseNova U1.5 Lite
  2. 2GPT Image 2 vs. SenseNova U1.5 Lite
  3. 3Reve 2.1 vs. SenseNova U1.5 Lite
  4. 4Nano Banana 2 vs. SenseNova U1.5 Lite
  5. 5GPT Image 1.5 vs. SenseNova U1.5 Lite
  6. 6MAI-Image-2.5 vs. SenseNova U1.5 Lite
  7. 7Nano Banana Pro vs. SenseNova U1.5 Lite
  8. 8Nano Banana 2 Lite vs. SenseNova U1.5 Lite
  9. 9Seedream 5.0 Pro vs. SenseNova U1.5 Lite
Show all 107 comparisons
  1. 10Grok Imagine Image Quality vs. SenseNova U1.5 Lite
  2. 11FLUX.2 Max vs. SenseNova U1.5 Lite
  3. 12Qwen Image 2.0 Pro vs. SenseNova U1.5 Lite
  4. 13MAI-Image-2.5 Flash vs. SenseNova U1.5 Lite
  5. 14HiDream-O1-Image-1.5 vs. SenseNova U1.5 Lite
  6. 15Krea 2 Medium Turbo vs. SenseNova U1.5 Lite
  7. 16Seedream 4.0 vs. SenseNova U1.5 Lite
  8. 17Krea 2 Large vs. SenseNova U1.5 Lite
  9. 18Luma Uni-1.1 Max vs. SenseNova U1.5 Lite
  10. 19FLUX.2 Flex vs. SenseNova U1.5 Lite
  11. 20Recraft V4.1 Utility vs. SenseNova U1.5 Lite
  12. 21Ideogram 4.0 vs. SenseNova U1.5 Lite
  13. 22Wan 2.6 Text to Image vs. SenseNova U1.5 Lite
  14. 23FLUX.2 Pro vs. SenseNova U1.5 Lite
  15. 24HunyuanImage 3.0 Instruct vs. SenseNova U1.5 Lite
  16. 25FLUX.2 Klein 9B vs. SenseNova U1.5 Lite
  17. 26HiDream-O1-Image vs. SenseNova U1.5 Lite
  18. 27Firefly Image Model 5 vs. SenseNova U1.5 Lite
  19. 28Midjourney V8.1 vs. SenseNova U1.5 Lite
  20. 29Stable Diffusion 3.5 Large vs. SenseNova U1.5 Lite
  21. 30GPT Image 2.5 Flare vs. SenseNova U1.5 Lite
  22. 31Qwen-Image-2.1 vs. SenseNova U1.5 Lite
  23. 32MAI-Image-2.6 vs. SenseNova U1.5 Lite
  24. 33MAI-Image-2.6 Flash vs. SenseNova U1.5 Lite
  25. 34Muse Image vs. SenseNova U1.5 Lite
  26. 35Grok Imagine Image 2.0 vs. SenseNova U1.5 Lite
  27. 36FLUX 3 Image vs. SenseNova U1.5 Lite
  28. 37FLUX 3 Image Klein vs. SenseNova U1.5 Lite
  29. 38Nano Banana 2.1 vs. SenseNova U1.5 Lite
  30. 39Ideogram 4.5 vs. SenseNova U1.5 Lite
  31. 40Hy Image 3.5 Preview vs. SenseNova U1.5 Lite
  32. 41Recraft V4.1 vs. SenseNova U1.5 Lite
  33. 42Recraft V4.1 Flash vs. SenseNova U1.5 Lite
  34. 43Bria FIBO 2 vs. SenseNova U1.5 Lite
  35. 44Krea 3 vs. SenseNova U1.5 Lite
  36. 45Qwen Image 3.0 Pro vs. SenseNova U1.5 Lite
  37. 46Qwen Image 3.0 vs. SenseNova U1.5 Lite
  38. 47Qwen Image 2.1 Pro vs. SenseNova U1.5 Lite
  39. 48Qwen Image 2.1 Turbo vs. SenseNova U1.5 Lite
  40. 49Wan 2.7 Image Pro vs. SenseNova U1.5 Lite
  41. 50Wan 2.7 Image vs. SenseNova U1.5 Lite
  42. 51Seedream 5.0 Flash vs. SenseNova U1.5 Lite
  43. 52SenseNova U1.5 Lite vs. Kling Image 3.0
  44. 53SenseNova U1.5 Lite vs. Kling Image 3.0 Omni
  45. 54SenseNova U1.5 Lite vs. Vidu Q2 Image
  46. 55SenseNova U1.5 Lite vs. GLM-Image
  47. 56SenseNova U1.5 Lite vs. MiniMax Image-01
  48. 57SenseNova U1.5 Lite vs. Z-Image Turbo
  49. 58SenseNova U1.5 Lite vs. Luma Uni-1.1
  50. 59SenseNova U1.5 Lite vs. Runway Gen-4 Image
  51. 60SenseNova U1.5 Lite vs. Adobe Firefly Image 3
  52. 61SenseNova U1.5 Lite vs. Adobe Firefly Image 4
  53. 62SenseNova U1.5 Lite vs. Adobe Firefly Image 4 Ultra
  54. 63SenseNova U1.5 Lite vs. Stable Diffusion 3.5 Medium
  55. 64SenseNova U1.5 Lite vs. Stable Diffusion 3.5 Large Turbo
  56. 65SenseNova U1.5 Lite vs. Bria FIBO Lite
  57. 66SenseNova U1.5 Lite vs. Recraft V4.1 Utility Pro
  58. 67SenseNova U1.5 Lite vs. Seedream 4.5
  59. 68SenseNova U1.5 Lite vs. Seedream 5.0 Lite
  60. 69SenseNova U1.5 Lite vs. FLUX.2 Klein 9B Preview
  61. 70SenseNova U1.5 Lite vs. GPT Image 1 Mini
  62. 71SenseNova U1.5 Lite vs. ChatGPT Image Latest (API-Alias)
  63. 72SenseNova U1.5 Lite vs. Nano Banana (Gemini 2.5 Flash Image)
  64. 73SenseNova U1.5 Lite vs. FLUX.2 Klein 4B
  65. 74SenseNova U1.5 Lite vs. FLUX.2 Dev
  66. 75SenseNova U1.5 Lite vs. FLUX1.1 [pro]
  67. 76SenseNova U1.5 Lite vs. FLUX1.1 [pro] Ultra
  68. 77SenseNova U1.5 Lite vs. FLUX.1 [dev]
  69. 78SenseNova U1.5 Lite vs. FLUX.1 Kontext [pro]
  70. 79SenseNova U1.5 Lite vs. FLUX.1 Kontext [max]
  71. 80SenseNova U1.5 Lite vs. Stable Diffusion XL 1.0
  72. 81SenseNova U1.5 Lite vs. Stable Diffusion 3 Medium
  73. 82SenseNova U1.5 Lite vs. SDXL Turbo
  74. 83SenseNova U1.5 Lite vs. Stable Diffusion Turbo
  75. 84SenseNova U1.5 Lite vs. Grok Imagine Image 1.0
  76. 85SenseNova U1.5 Lite vs. MOSS-Image-Base
  77. 86SenseNova U1.5 Lite vs. MOSS-Image-SFT
  78. 87SenseNova U1.5 Lite vs. LLaDA-Image
  79. 88SenseNova U1.5 Lite vs. LLaDA-Image-Turbo
  80. 89SenseNova U1.5 Lite vs. Nucleus-Image
  81. 90SenseNova U1.5 Lite vs. LongCat-Image
  82. 91SenseNova U1.5 Lite vs. BAGEL-7B-MoT
  83. 92SenseNova U1.5 Lite vs. OmniGen2
  84. 93SenseNova U1.5 Lite vs. Kandinsky-5.0-T2I-Lite-SFT
  85. 94SenseNova U1.5 Lite vs. Kandinsky-5.0-T2I-Lite-Pretrain
  86. 95SenseNova U1.5 Lite vs. Boogu-Image-0.1-Base
  87. 96SenseNova U1.5 Lite vs. Boogu-Image-0.1-Turbo
  88. 97SenseNova U1.5 Lite vs. SANA-1.5 1.6B 1024px
  89. 98SenseNova U1.5 Lite vs. SANA-1.5 4.8B 1024px
  90. 99SenseNova U1.5 Lite vs. Sana-Sprint 0.6B 1024px
  91. 100SenseNova U1.5 Lite vs. Sana-Sprint 1.6B 1024px
  92. 101SenseNova U1.5 Lite vs. Chroma1-HD
  93. 102SenseNova U1.5 Lite vs. Chroma1-Flash
  94. 103SenseNova U1.5 Lite vs. AuraFlow v0.3
  95. 104SenseNova U1.5 Lite vs. PixArt-Sigma-XL-2-1024-MS
  96. 105SenseNova U1.5 Lite vs. Lumina-Image 2.0
  97. 106SenseNova U1.5 Lite vs. HunyuanImage 3.0
  98. 107SenseNova U1.5 Lite vs. Z-Image Base

All specifications

Input and output

Maximum output
4K
Source: Maximum output

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Unknown
no source
Watermarking technique
Not known
no source

Licence and access

Weights
Open weights
Source: Weights
Access
API, Web, Local
Source: Access

API

Native API model ID
sensenova-u1.5-lite; available as an image model through the SenseNova Token Plan.
Source: Native API model ID

Further details

Official weights
The Lite announcement links SenseNova-U1.5-8B-MoT. The official weights are available under Apache-2.0.
Source: Official weights
Weight checkpoint
sensenova/SenseNova-U1.5-8B-MoT
Source: Weight checkpoint
Access detail
The live vendor page dates Token Plan availability to August 25, 2026, although its URL path says September 11, 2026.
Source: Access detail
Prompt processing
The API can expand prompts automatically with prompt_extend; disable it for a controlled comparison if the API permits.
Source: Prompt processing

Frequently asked questions about SenseNova U1.5 Lite

How good is SenseNova U1.5 Lite?

Rank 14 of 108 in our own test.

GraduallyBench from Nov 3, 2026

GraduallyBenchour own rating, 0 to 100 · higher is better
FLUX.2 Flex73
Grok Imagine Image 1.073
Reve 2.173
SenseNova U1.5 LiteRank 14 of 10873
GPT Image 273
Grok Imagine Image 2.073
Qwen Image 2.0 Pro72

What is SenseNova U1.5 Lite good at?

Best at Typography and layout (93 out of 100), weakest at Character and detail (35).

GraduallyBench, one image per task

  1. SenseNova U1.5 Lite: Typography and layoutTypography and layoutLegible text, hierarchy, and composition93out of 100Rank 6 of 108 in this task
  2. SenseNova U1.5 Lite: Product photographyProduct photographyMaterials, reflections, and fine details84out of 100Rank 18 of 108 in this task
  3. SenseNova U1.5 Lite: Character and detailCharacter and detailAnatomy, clothing, and spatial consistency35out of 100Rank 81 of 108 in this task
  4. SenseNova U1.5 Lite: InfographicInfographicNumbers, labels, and visual organization80out of 100Rank 15 of 108 in this task

Sources and data date

Show all 10 sources

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.

Without vendor claims and historical or disputed results; those stay labelled in the single values.