SenseNova-U1.5-8B-MoT-Preview

SenseNova-U1.5-8B-MoT (Preview)

English 简体中文

• *NOTE:** SenseNova-U1.5 (preview) offers an early look at our latest progress in image generation and editing. Notably, a more capable and refined production version will be officially released soon.

🌟 Overview

SenseNova-U1.5 (Preview)** is a natively unified multimodal model built on

NEO-unify, featuring new patch encoding and decoding layers. We further expand and refine the text-to-image training corpus, while extensively filtering and synthesizing editing data for both single-image and multi-image reference settings. Hence, it shows a notable leap in native image generation and editing:

• *Native 4K generation with remarkable efficiency**, preserving both global coherence and fine-grained detail at high resolution.
• *Striking improvements in texture, material, lighting, and realism**, producing richer and more visually convincing images.
• *Substantially stronger Chinese and English text rendering**, together with more reliable composition of dense and complex layouts.
• *A major step forward in image editing**, with stronger instruction following, subject-identity preservation, and structural consistency.
• *Fine-grained, region-controllable editing** through masks, bounding boxes, and visual markers.

📣 Updated News

[2026.07.31] Released **SenseNova-U1.5-8B-MoT (Preview)**, featuring native 4K generation, finer textures, complex layouts, and stronger editing capability. We also open-sourced the pre-training launcher and configuration.

Click to expand the SenseNova U1 updated history

[2026.07.16] Released SenseNova-U1-8B-MoT-Infographic-V3 📊. The new version supports both infographic generation and editing. While retaining its infographic generation capabilities, it substantially improves local text, local content, global style, and global layout editing, including precise text correction in dense text. For model details and benchmark results, see ✨ U1 Infographic Model Series.

[2026.06.29] Released SenseNova-U1-8B-MoT-Infographic-V2 📊. This version improves dense small-text rendering with sharper character edges, strengthens layout generation for complex and information-dense graphics, improves overall visual quality and coherence, and fixes the black-background issue. For model details and visual examples, see ✨ U1 Infographic Model Series.

[2026.06.12] Released SenseNova-U1-8B-MoT-Infographic-LoRA-8step-V1.0 for fast infographic generation. See the inference examples.

[2026.06.11] Released SenseNova-U1-8B-MoT-Interleaved 📖, optimized specifically for interleaved text-image generation. It substantially improves narrative coherence, character and style consistency, and text-image alignment across multi-page content such as picture books, storybooks, slide decks, and illustrated tutorials.

[2026.05.21] Released the full-parameter fine-tuning training code for SenseNova U1.

[2026.05.15] Released SenseNova-U1-8B-MoT-Infographic 📊, improving infographic generation. For model details, see U1 Infographic Model Series; for 100 generation examples, see ✨ Infographic Showcases.

[2026.05.10] Released the 🔥 SenseNova U1 Technical Report 🔥 and open-sourced the SenseNova-U1-A3B-MoT-SFT and SenseNova-U1-A3B-MoT model weights.

[2026.05.08] Added **GGUF quantized-weight support** and a **layered-loading VRAM mode** for inference on a single GPU with limited memory. See the low-VRAM inference guide. GGUF weights for SenseNova-U1-8B-MoT-Merger are available at 🤗 smthem/SenseNova-U1-8B-MoT-Merger-gguf. We thank @smthem for contributing the quantized weights to the community.

[2026.05.06] Released SenseNova-U1-8B-MoT-LoRA-8step-V1.0. See the inference examples.

[2026.04.30] Released the 8-step inference preview SenseNova-U1-8B-MoT-8step-preview. In most cases, its image-generation quality is close to that of the base model. See the comparison and known issues. At runtime, follow the inference scripts and use --cfg scale 1.0 --num steps 8 .

[2026.04.27] First released the SenseNova-U1-8B-MoT-SFT and SenseNova-U1-8B-MoT model weights.

[2026.04.27] First released the SenseNova U1 inference code.

🔬 What U1.5 Improves

From Patchwise Prediction to Patch-Joint Reconstruction

SenseNova-U1's original MLP head reconstructs each RGB patch independently. At high resolutions, this can expose token boundaries as grid patterns, seams, or texture discontinuities. SenseNova-U1.5 instead adopts progressive spatial reconstruction via ConvDecoder: visual tokens are reshaped into a two-dimensional feature grid and upsampled through multiple Pixel Shuffle stages, with intermediate 3×3 convolutions enabling neighboring patches to interact and merge into a more continuous image.

Qualitative results show that this design substantially suppresses grid-like artifacts throughout pre-training. It also improves robustness during downstream adaptation, reducing the likelihood that similar artifacts re-emerge when fine-tuning on new domains.*

From Instruction Following to Visual Preservation

SenseNova-U1 relied primarily on publicly available image-editing datasets, many of which contained noisy supervision, low-quality samples, and visible compositing or retouching artifacts. For SenseNova-U1.5, we extensively clean, filter, and synthesize the editing corpus, while improving Chinese–English balance and expanding coverage of both single-image and multi-image reference settings.

Quantitative results show that the encoder-free NEO-unify architecture achieves strong instruction following together with subject-identity and structural preservation using only a single sequence of reference-image tokens. This results in stronger editing performance and substantially higher inference efficiency, further validating the effectiveness of native unified multimodal modeling.*

From Format Exposure to Cross-Task Generalization

The generation and editing corpora used to train SenseNova-U1.5 contain only a small number of simple JSON-formatted prompts, rather than long, hierarchical, or highly structured instructions. Besides, more complex JSON-like formats appear only in the understanding training data corpora, with substantially different content, structures and task objectives.

Despite this gap, it generalizes well to long and structured instructions in image generation and editing. This suggests that native unified multimodal modeling can combine capabilities learned from understanding and generation within a unified model, enabling structured comprehension and visual planning to transfer across tasks without dedicated training on fixed prompt templates.*

✨ What U1.5 Achieves

🖼️ Text-to-Image

Qwen-Image Bench

Model

Quality

Aesthetics

Alignment

Real-world Fidelity

Creative Generation

Overall

EN ↑

ZH ↑

EN ↑

ZH ↑

EN ↑

ZH ↑

EN ↑

ZH ↑

EN ↑

ZH ↑

EN ↑

ZH ↑

GLM Image

• 8649.26
• 9850.64
• 4947.90
• 2544.69
• 6745.23
• 8648.19

Qwen Image

• 4548.44
• 1852.25
• 0450.72
• 4543.16
• 3747.30
• 4849.23

HunyuanImage 3.0

• 7650.35
• 6653.57
• 1652.00
• 3344.31
• 3349.12
• 3550.81

Qwen Image 2512

• 8451.76
• 4054.74
• 4452.72
• 8047.00
• 7550.19
• 3252.06

GPT Image 1

• 5052.34
• 7755.09
• 3556.28
• 2548.14
• 2955.78
• 2454.07

Qwen Image 2*

• 6053.09
• 5156.30
• 4857.48
• 1751.74
• 4059.01
• 6955.63

GPT Image 2

• 0958.65
• 4867.53
• 7865.85
• 4057.38
• 3475.23
• 2364.69

U1

• 3045.79
• 3047.07
• 5847.78
• 1342.53
• 1543.88
• 2845.99

U1.5-Preview

• 1449.19
• 0251.87
• 4652.61
• 2245.11
• 2348.80
• 9350.25

U1.5-Preview†

• 2150.01
• 2656.44
• 5259.28
• 2648.24
• 4360.95
• 1755.22

Note: Qwen Image 2* denotes our evaluation results; U1.5-Preview† denotes results obtained using Prompt Enhance (PE), implemented following the Cosmos3 upsampling strategy..

✏️ Image Editing

Model

ImgEdit-Bench

GEdit-Bench

WeEdit

Overall ↑

EN GO ↑

CN GO ↑

IA ↑

TC ↑

BP ↑

Average ↑

Qwen-Image-Edit-2511

• 51
• 877
• 819
• 180
• 930
• 630
• 913

Qwen-Image-2

• 28
• 369
• 348
• 044
• 079
• 680
• 601

FireRed-Image-Edit

• 56
• 943
• 887
• 150
• 330
• 140
• 873

LongCat-Image-Edit

• 45
• 748
• 731
• 390
• 590
• 710
• 230

HY-Image-3-Instruct

—

—

—

• 160
• 990
• 030
• 727

Nano-Banana

• 29
• 291
• 399
• 920
• 140
• 800
• 287

Nano-Banana-Pro

• 37
• 738
• 799
• 580
• 100
• 850
• 843

GPT-Image-1.5

—

—

—

• 520
• 780
• 150
• 817

U1

• 90
• 470
• 420
• 729
• 604
• 157
• 497

U1.5-Preview

• 37
• 172
• 051
• 532
• 271
• 752
• 852

🎨 Showcases

🖼️ Text-to-Image Showcase

The following examples cover several types of text-to-image tasks. Click a title

in the left column to expand the complete original prompt.

Prompt

SenseNova-U1.5 (Preview)

Vintage soda advertisement

生成一张竖版复古可乐商业广告海报，品牌为“RETRO-COLA”，画面比例约为 2:3。整体采用 1960—1970 年代美式汽水广告、复古泳池派对摄影、旧杂志印刷和现代商业时尚广告融合的风格。画面必须真实、热烈、复古、富有生活方式气息，不能呈现为插画、卡通、漫画或明显的三维渲染。

场景设定在夏夜的户外露台或泳池派对。背景为深靛蓝色夜空，远处有少量暖黄色虚化灯光和轻微模糊的城市或派对环境。画面顶部撑开一把大型复古遮阳伞，伞面是奶油黄与焦糖橙相间的宽条纹，带真实布料褶皱、纤维质感和边缘流苏。遮阳伞占据画面顶部区域，但不能压住主要文字。

画面右侧中央是一位年轻女性，她是整张海报的人物主角。她有浓密蓬松的深棕色短卷发，佩戴金色圆环耳环、细项链和手镯，皮肤具有真实细腻的暖棕色质感。她表情自然、开怀大笑，充满夏夜派对的放松与感染力，像真实抓拍而不是僵硬摆拍。

她穿一件黄色与奶油白色格纹的复古修身连体泳装或复古吊带连衣装，胸口为复古收褶结构，肩带清楚，服装贴合身体但自然不过度夸张。人物采用稳定清晰的三分之四正面坐姿，稳定地坐在一个装满透明冰块的大型黄色金属冰桶里。上半身完整可见，头部、颈部、双肩、双臂和躯干结构清楚。她的背部自然挺直，肩膀放松，左右肩高度合理，胸腔和腰部略微朝向镜头，人体比例真实。

肢体结构必须清楚、简洁、稳定。画面中只保留主角自己的一双手，不要出现第三只手、额外手臂或从镜头外伸入的其他人的手。她的右手握住一瓶玻璃瓶装可乐，将瓶子自然举在胸口附近。右上臂贴近身体，手肘自然弯曲，前臂方向清楚。右手必须完整呈现五根手指，拇指在瓶身一侧，其余四指自然环绕瓶身，手指长度、关节和弯曲方向真实，不能粘连、穿模、扭曲，也不能遮挡主要标签文字。左手自然搭在黄色冰桶边缘，手掌放松，五根手指轻轻弯曲并清楚分开，手与桶沿有真实接触关系和轻微阴影，不隐藏在身体后方，也不与冰块、衣服或腿部融合。双臂不能交叉，双手不能遮挡脸部，也不能互相重叠。

腿部也要保持简洁稳定。人物只露出一条自然弯曲抬起的腿，只显示一个清楚的膝盖和部分大腿，膝盖位于画面中央偏右，形状自然，和躯干连接合理。另一条腿完全隐藏在冰桶和冰块内部，不要额外显示第二个膝盖、脚、小腿或重复腿部。冰桶边缘明确遮挡腰部以下位置，遮挡关系自然。不要采用双腿交叉、复杂盘腿、扭曲髋部或不可能的坐姿。

她右手拿着的可乐瓶必须保持真实产品结构：透明深棕色玻璃瓶、细长瓶颈、金属瓶盖、合理玻璃厚度和真实折射。瓶内装深褐色碳酸饮料，液面自然，内部有少量细密气泡，瓶外覆盖冷凝水珠。瓶身红棕色纸质标签上清晰显示品牌文字：

“RETRO-COLA”

使用奶油白色复古粗衬线字体。标签紧贴瓶身曲面，不悬浮、不变形、不重复、不出现乱码。

画面底部中央是一只大型复古黄色金属冰桶，桶壁带真实冷凝水珠、少量磨损和金属反光。桶中堆满不规则透明冰块，冰块有自然裂纹、融水和高光。桶身正面印刷大型红棕色品牌文字：

“DRINK”

“RETRO-COLA”

其中“RETRO-COLA”使用复古弧形衬线字体，带少量星芒装饰，文字要清晰贴合桶体透视。

画面前景左右下角各放置一瓶靠近镜头的 RETRO-COLA 玻璃瓶。两瓶作为前景景深元素，被边缘轻微裁切，并略有前景虚化，但仍然能够辨认出深色玻璃、红棕标签和奶油白品牌文字。除了这两瓶前景产品，不要再加入额外人物手臂，不要有左下角伸手的动作元素。

画面左侧约占四成宽度，放置五行超大黄色英文标题：

“VINTAGE”

“CARBONATED”

“DREAMS!”

“UNCAP”

“THE PAST.”

文字从上到下排列，使用高窄、极粗、略带手工印刷感的无衬线字体，颜色为芥末黄色。五行左对齐、行距紧凑、占据强烈视觉冲击力。标题位于人物左侧的深蓝夜空区域，不覆盖人物面部，不与可乐瓶标签重叠，每一行都必须清晰可读。

顶部中央放置较小黑色标题：

“THE RETRO FIZZ®”

左上角放置三行黄色小字：

“FUTURE”

“RETRO”

“FLAVOR”

右上角放置两行黄色小字：

“RETRO-COLA”

“LIFE 2026”

所有顶部小字均使用窄体粗无衬线字体，紧凑清晰，保持克制，不抢主体。

光线采用夜间派对中的暖色直接闪光灯或暖色商业布光与环境夜色结合。人物、玻璃瓶、冰块和黄色冰桶被温暖而清楚地照亮，皮肤、头发、玻璃瓶和冰块高光真实。背景保持深靛蓝色夜景和少量暖色散景。画面可以带适量复古胶片颗粒、轻微油墨噪点和旧杂志印刷感，但不能牺牲人物面部、手部、产品包装和文字清晰度。

整体配色以深靛蓝、芥末黄、焦糖橙、暖棕色、奶油白和可乐红棕色为主。整体气质像真实拍摄的复古汽水品牌广告，可用于饮料新品发布、夏季派对活动、电商主视觉和社交媒体营销。

特别强调：

主角只有一双手；右手握瓶，左手搭在冰桶边缘；只露出一个膝盖；另一条腿完全藏在冰桶中；不出现额外手臂、额外手指、重复膝盖、扭曲坐姿或肢体融合。

Marine turtle educational infographic

{

"type": "circular life-cycle educational infographic poster",

"theme": "MARINE TURTLES of the MALDIVES - Life Cycle and Threats",

"language": "English",

"priority emphasis": [

"TOP PRIORITY: all text crisp, correctly spelled, plain letters and numbers only - no strange symbols or garbled characters",

"core structure: a circular five-stage life cycle arranged clockwise around a central dark-teal oval label, with curved arrows connecting the stages",

"style: soft watercolor natural-history illustration inside each stage circle, combined with clean flat icons and elegant serif typography",

"calm coastal palette: soft ocean teals and blues on a very light blue-white background, one warm sand-beige accent band"

],

"canvas": {

"aspect ratio": "2:3",

"orientation": "vertical poster",

"resolution hint": "high resolution, refined print-quality detail",

"background": "very light blue-white with a subtle watercolor wash texture"

},

"style": {

"overall": "elegant marine conservation poster - watercolor illustration meets clean infographic design",

"visual tone": "calm, natural, scientific, gentle",

"art direction": "each life stage is a circular watercolor vignette with soft edges; around each vignette a pale-teal arc segmented into small flat threat icons; bold curved teal arrows link the stages clockwise; dotted lines connect stages to the center",

"color palette": [

"deep ocean teal (headings, arrows, icons)",

"soft aqua and powder blue (arcs and fills)",

"muted seafoam green",

"warm sand beige (bottom accent band)",

"very light blue-white background"

],

"saturation": "low to medium, watercolor softness, no bright colors",

"typeface": "classic serif in dark teal - large elegant serif for the title with an italic word, medium serif for stage names and section headers, clean sans-serif regular for body bullets",

"references": ["natural history watercolor plates", "marine conservation posters", "editorial life-cycle diagrams"]

},

"layout": {

"header": {

"title": "MARINE TURTLES of the MALDIVES - large dark-teal serif, the words of the in elegant italic",

"subtitle": "Life Cycle and Threats - medium serif flanked by small decorative dashes"

},

"central hub": "a solid dark-teal oval in the middle of the cycle with white italic serif text: Life Cycle and Threats",

"life cycle ring": {

"arrangement": "five stages arranged clockwise: stage 1 top center, stage 2 right, stage 3 lower right, stage 4 lower left, stage 5 left; bold curved teal arrows between consecutive stages; thin dotted lines from each stage toward the central oval",

"stage design": "each stage has a numbered dark-teal circle badge on top, a round watercolor illustration, a serif stage name below, and a pale-teal arc of 4-5 small flat threat icons hugging the outer side of the circle (crab, bird, fish, thermometer, net, plastic bottle, no-poaching sign, excavator, boat as appropriate)",

"stages": [

{"number": "1", "name": "Eggs and Nest", "illustration": "watercolor female turtle on warm sand laying a clutch of white eggs among beach grass", "side text": "left column bullets: Females return to natal Maldivian beaches to nest. Clutch approx 80-120 eggs; incubation approx 45-70 days in the warm sand.", "right text": "right column bullets: Bycatch in fishing gear and ghost-net entanglement. Plastic ingestion, poaching, coastal development and boat strikes."},

{"number": "2", "name": "Hatchlings", "illustration": "watercolor night beach scene with a tiny hatchling scrambling toward moonlit waves", "side text": "bullets: Hatchlings emerge at night and scramble to the sea toward the brightest horizon. Survival to adulthood is estimated at only approx 1 in 1,000."},

{"number": "3", "name": "The Lost Years", "illustration": "watercolor juvenile turtle drifting among Sargassum seaweed and small fish in open blue water", "side text": "bullets: Juveniles drift on open-ocean currents and Sargassum for years. This long pelagic phase is known as the lost years."},

{"number": "4", "name": "Sub-adult on Reefs", "illustration": "watercolor young turtle swimming over coral reef and seagrass", "side text": "bullets: Young turtles recruit to Maldivian coral reefs and seagrass beds. They feed and grow toward maturity in coastal foraging grounds."},

{"number": "5", "name": "Adult and Migration", "illustration": "watercolor mature turtle swimming through open water with light rays", "side text": "bullets: Mature adults migrate between foraging and breeding grounds. Sexual maturity is reached late - often 20-30 years."}

]

},

"species status row": {

"left": "a watercolor hawksbill turtle illustration beside header Maldivian Species with bullets: Five of the seven sea-turtle species occur in Maldivian waters. Green and Hawksbill turtles are the most commonly seen.",

"right": "a dark-teal shield icon with a white turtle silhouette beside header Conservation Status with bullets: Hawksbill Critically Endangered. Green Endangered (IUCN Red List). A second watercolor green turtle illustration at the far right."

},

"threat bands": {

"natural": "a full-width pale-teal rounded band titled NATURAL THREATS with four icon-text pairs separated by dotted dividers: crab icon - Predation of eggs and hatchlings by crabs, birds and fish; virus icon - Disease; coral icon - Coral and seagrass loss; storm cloud icon - Storms erode nesting beaches.",

"human": "below it a full-width warm sand-beige rounded band titled HUMAN THREATS with five icon-text pairs: fishing net - Bycatch in fishing gear and ghost-net entanglement; plastic bottles - Plastic ingestion; turtle with hand - Poaching; excavator - Coastal development; boat - Boat strikes."

}

},

"composition rules": [

"the five-stage ring is the dominant structure, evenly spaced around the central oval",

"watercolor vignettes share a consistent illustration style and scale",

"threat-icon arcs hug each stage circle on its outer side without touching neighboring elements",

"side text columns sit outside the ring, aligned left or right, never overlapping the cycle",

"curved arrows are bold and smooth, clearly showing clockwise flow",

"the two threat bands stretch the full width at the bottom with even icon spacing",

"generous breathing room, nothing touches the poster edges"

],

"quality tags": [

"watercolor natural-history illustration",

"circular life-cycle diagram",

"marine conservation poster",

"soft coastal palette",

"elegant serif typography",

"clean flat icon system",

"crisp accurate text",

"high detail"

],

"negative prompt": [

"garbled text",

"misspelled words",

"strange symbols",

"blurry small labels",

"bright neon colors",

"oversaturated palette",

"photorealistic turtles",

"3D render look",

"cartoon childis