Views
No views yet
[2026.05.10] Release 🔥SenseNova-U1 Technical Report🔥 and the weights for SenseNova-U1-A3B-MoT-SFT & SenseNova-U1-A3B-MoT.[2026.05.08] Add GGUF quantized checkpoints and layer-offload VRAM modes for low-VRAM single-GPU inference. See Memory-efficient inference. GGUF weights for SenseNova-U1-8B-MoT-Merger are available at 🤗 smthem/SenseNova-U1-8B-MoT-Merger-gguf — many thanks to @smthem for contributing the quantized weights.[2026.04.30] Release the preview version of the 8-step inference model SenseNova-U1-8B-MoT-8step-preview. In most cases, the image generation quality of this model closely matches that of the base model (see comparison and existing issues). To test this model, you can use the inference scripts, but with the following parameters: --cfg_scale 1.0 --num_steps 8 .[2026.04.27] Initial release of the inference code for SenseNova-U1.
 |  |
| Model | Params | HF Weights |
|---|---|---|
| SenseNova-U1-8B-MoT-SFT | 8B MoT | 🤗 link |
| SenseNova-U1-8B-MoT | 8B MoT | 🤗 link |
| SenseNova-U1-8B-MoT-LoRA-8step-V1.0 | 0.4B | 🤗 link |
| SenseNova-U1-A3B-MoT-SFT | A3B MoT | 🤗 link |
| SenseNova-U1-A3B-MoT | A3B MoT | 🤗 link |
💡 The8B-MoTinSenseNova-U1-8B-MoTrefers to ~8B understanding parameters and ~8B generation parameters. See parameter breakdown for details.
| Original Text | Reasoning Process | Resulting Image |
|---|---|---|
| A male peacock trying to attract a female | 1. Instruction Understanding: The core subject is a single male peacock. The action involves the male peacock trying to attract a female. 2. Reasoning Process: Male peacocks perform this ritual by fanning their tail feathers into a massive, colorful circular display. 3. Establish the frame: The frame captures the male peacock in a full-body shot to show the entire display. The angle is eye-level to emphasize the height of the tail feathers. 4. Set the lighting and color: Natural lighting illuminates the colorful circular display with soft shadows. The palette features vibrant blues and greens on the tail feathers. 5. Lock the style: The style is a realistic photograph with sharp focus on the male peacock. The finish is natural without artistic filters. 6. Explicit Prompt: A vibrant male peacock with his tail feathers fully fanned out in a wide, iridescent display of blues and greens, standing in a garden. |  |
| A small piece of dry wood and a dense iron block are in a transparent water tank | 1. Instruction Understanding: The core subjects are a small piece of dry wood and a dense iron block positioned within a transparent water tank. The wood is floating on the surface while the iron block is submerged at the bottom. 2. Reasoning Process: The wood is less dense than water so it will float, while the iron is denser and will sink to the bottom. 3. Establish the frame: The composition is a vertical medium shot centering the rectangular tank within the frame. The camera angle is eye-level to clearly display the water line and the submerged base. Focus is sharp across the entire depth of the tank to ensure both materials are distinct. 4. Build the environment: The scene is contained entirely within the clear glass walls of the water tank. The water fills the majority of the volume, providing a medium for the floating wood and sunken iron block. The background remains out of focus to keep attention on the tank's interior. 5. Set the lighting and color: Soft natural light illuminates the scene from the left, creating gentle reflections on the water surface. The color palette features the brown grain of the wood contrasting against the dark grey metallic finish of the iron. Shadows are soft and diffused through the liquid. 6. Explicit Prompt: A realistic photo of a transparent water tank showing a piece of wood floating on the surface and an iron block resting at the bottom. |  |
| A typical dish from the country where Naples is located | 1. Instruction Understanding: The core subject is a Neapolitan pizza presented as a typical dish. The context specifies Naples, Italy, as the country of origin for this food item. 2. Reasoning Process: Naples is in Italy, and a classic dish is a Neapolitan pizza. 3. Establish the frame: The Neapolitan pizza is captured in a close-up shot that fills the central frame. The angle is slightly elevated to show the round form of the dish clearly. 4. Set the lighting and color: Soft lighting illuminates the surface of the Neapolitan pizza to reveal texture. Warm tones dominate the color palette, emphasizing the baked nature of the dish. 5. Lock the style: The image utilizes a realistic photographic style with sharp focus on the main subject. The finish appears natural and appetizing, suitable for food documentation. 6. Explicit Prompt: A delicious Neapolitan pizza with a soft, charred crust, tomato sauce, and fresh mozzarella, served on a rustic wooden table, realistic food photography. |  |
| A solution of calcium carbonate reacting with acetic acid | 1. Instruction Understanding: The core subject is a solution of calcium carbonate and acetic acid. The prompt specifies the reacting state of the chemical mixture. 2. Reasoning Process: The reaction produces carbon dioxide gas, which would be visible as a steady stream of bubbles rising through the liquid. 3. Establish the frame: The camera frames the solution closely to capture the details of the reaction. The composition centers on the liquid where the gas is visible. 4. Set the lighting and color: The liquid appears clear, allowing the white bubbles to stand out distinctly. The lighting is bright and even to illuminate the stream of gas. 5. Lock the style: The image maintains a realistic photographic style suitable for scientific observation. The focus is sharp on the reacting solution and bubbles. 6. Explicit Prompt: A test tube filled with a clear liquid and a rapid, effervescent stream of carbon dioxide bubbles rising to the surface, laboratory experiment. |  |
📸 More generation samples: see Image Generation Gallery.
  Replace the text "WARFIGHTER" to "BATTLEFIELD" in the bold orange-red font. |
| Original Text | Original Image | Reasoning Process | Resulting Image |
|---|---|---|---|
| Draw what it will look like one hour later. |  |
1. Source Image Analysis: The source image shows a glass cup of hot tea with steeping tea leaves, and the water appears relatively clear. 2. Instruction Understanding: The instruction asks for the appearance one hour later. 3. Reasoning Process: Over time, tannins and pigments leach out, making the tea noticeably darker and more uniformly colored, and the leaves may look more swollen and darker from soaking. 4. Expected Visual Changes: The expected visible result is a deeper amber-to-brown tea color and more fully saturated liquid. 5. Elements to Preserve: The glass cup, scattered leaves around it, background, and camera angle should remain unchanged. 6. Explicit Edit Prompt: Edit the tea liquid to a much darker, more saturated amber-brown color as if fully steeped, and make the tea leaves look slightly darker and more swollen, while keeping the glass cup, surrounding leaves, background, and framing unchanged. |  |
| Draw what it will look like immediately after someone stands up from sitting on it for a long time. |  |
1. Source Image Analysis: The source image shows a fluffy lime-green beanbag chair that looks evenly plump and undisturbed on a white background. 2. Instruction Understanding: The edit instruction asks for its appearance immediately after someone stood up from sitting there for a long time. 3. Reasoning Process: Prolonged weight compresses the fabric and internal fill, leaving a depressed seat area, wrinkles radiating outward, and a slowly recovering shape. 4. Expected Visual Changes: The visible result should be a noticeable dip and creasing where a person was seated. 5. Elements to Preserve: The background, beanbag color, lighting, and camera angle should remain unchanged while only the beanbag’s shape shows the compression. 6. Explicit Edit Prompt: Edit the beanbag chair to show a clear seated depression in the center with surrounding wrinkles and slightly compressed fabric, while keeping the white background, lighting, and camera angle unchanged. |  |
| Change the water to high-concentration saltwater |  |
1. Source Image Analysis: The source image shows an egg resting at the bottom of a glass of water. 2. Instruction Understanding: The instruction asks to change the medium to high-concentration saltwater. 3. Reasoning Process: Saltwater is denser than fresh water, which increases the buoyant force on the egg. 4. Expected Visual Changes: As density increases, the egg will overcome gravity and float higher or suspend in the middle of the liquid. 5. Elements to Preserve: The glass and the egg's appearance should remain consistent, focusing on the shift in the egg's vertical position. 6. Explicit Edit Prompt: Edit the position of the egg so it is floating in the middle of the liquid instead of resting on the bottom, while keeping the glass and the egg's appearance unchanged. |  |
| What the fruit looks like when ripe in the picture |  |
1. Source Image Analysis: The source image shows green, unripe bananas. 2. Instruction Understanding: The instruction asks for the appearance of the fruit when ripe. 3. Reasoning Process: Ripening involves a breakdown of chlorophyll and the production of sugars, which turns the skin from green to yellow and often causes small brown sugar spots to appear. 4. Expected Visual Changes: The color and texture of the peel should transition to a ripe state. 5. Elements to Preserve: The shape of the bananas and the white background should remain constant. 6. Explicit Edit Prompt: Edit the green bananas to be bright yellow with small brown spots, while keeping the original shape and white background unchanged. |  |
📸 More editing samples: see Image Editing Gallery.
📸 More interleaved samples: see Interleaved Generation Gallery.
📸 More understanding samples: see Visual Understanding Gallery.
Evaluation scripts and benchmark reproduction guides are added inevaluation.
prompt enhancement for best practice)Note: To serve more users, U1-Fast has undergone step and CFG distillation, and is dedicated to infographic generation.
Refer to the SenseNova-Skills README for installation and usage details.
Setup: Follow the Installation Guide to clone the repo and install dependencies with uv.
python examples/vqa/inference.py --model_path sensenova/SenseNova-U1-8B-MoT --image examples/vqa/data/images/menu.jpg --question "My friend and I are dining together tonight. Looking at this menu, can you recommend a good combination of dishes for 2 people? We want a balanced meal — a mix of mains and maybe a starter or dessert. Budget-conscious but want to try the highlights." --output outputs/answer.txt --max_new_tokens 8192 --do_sample --temperature 0.6 --top_p 0.95 --top_k 20 --repetition_penalty 1.05 --profileSeeexamples/README.mdfor batched inference, generation parameters, and JSONL format.
python examples/t2i/inference.py --model_path sensenova/SenseNova-U1-8B-MoT --prompt "这张信息图的标题是“SenseNova-U1”，采用现代极简科技矩阵风格。整体布局为水平三列网格结构，背景是带有极浅银灰色细密点阵的哑光纯白高级纸张纹理，画面长宽比为16:9。\n\n排版采用严谨的视觉层级：主标题使用粗体无衬线黑体字，正文使用清晰的现代等宽字体。配色方案极其克制，以纯白色为底，深炭黑为主视觉文字和边框，浅石板灰用于背景色块和次要信息区分，图标采用精致的银灰色线框绘制。\n\n在画面正上方居中位置，使用醒目的深炭黑粗体字排布着大标题“SenseNova-U1”。标题正下方是浅石板灰色的等宽字体副标题“新一代端到端统一多模态大模型家族”。\n\n画面主体分为左、中、右三个相等的垂直信息区块，区块之间通过充足的负空间进行物理隔离。\n\n左侧区块的主题是概述。顶部有一个银灰色线框绘制的、由放大镜和齿轮交织的图标，旁边是粗体小标题“Overview”。该区块内从上到下垂直排列着三个要点：第一个要点旁边是一个代表文档与照片重叠的极简图标，紧跟着文字“多模态模型家族，统一文本/图像理解和生成”。向下是由两个相连的同心圆组成的架构图标，配有文字“基于NEO-Unify架构（端到端统一理解和生成）”。最下方是一个带有斜线划掉的眼睛和漏斗形状的图标，明确指示文本“无需视觉编码器(VE)和变分自编码器(VAE)”。\n\n中间区块展示模型矩阵。顶部是一个包含两个分支节点的树状网络图标，旁边是粗体小标题“两个模型规格”。区块内分为上下两个包裹在浅石板灰色极细边框内的卡片。上方的卡片内画着一个代表高密度的实心几何立方体图标，大字标注“SenseNova-U1-8B-MoT”，下方是等宽字体说明“8B MoT 密集主干模型”。下方的卡片内画着一个带有闪电符号的网状发光大脑图标，大字标注“SenseNova-U1-A3B-MoT”，下方是等宽字体说明“A3B MoT 混合专家（MoE）主干模型”。在这两个独立卡片的正下方，左侧放置一个笑脸轮廓图标搭配文字“将在HF等平台公开”，右侧放置一个带有折角的书面报告图标搭配文字“将发布技术报告”。\n\n右侧区块呈现核心优势。顶部是一个代表巅峰的上升阶梯折线图图标，旁边是粗体小标题“Highlights”。该区块内部垂直分布着四个带有浅石板灰底色的长方形色块，每个色块内部左侧对应一个具体的图标，右侧为文字。第一个色块内是一个无缝相连的莫比乌斯环图标，配文“原生统一架构，无VE和VAE”。第二个色块内是一个顶端带有星星的奖杯图标，配文“单一统一模型在理解和生成任务上均达到SOTA性能”。第三个色块内是代表文本行与拍立得照片交替穿插的图标，配文“强大的原生交错推理能力（模型原生生成图像进行推理）”。最后一个色块内是一个被切分出一小块的硬币与详细饼状图结合的图标，配文“能生成复杂信息图表，性价比出色”。" --width 2720 --height 1536 --cfg_scale 4.0 --cfg_norm none --timestep_shift 3.0 --num_steps 50 --output output.png --profileDefault resolution is 2048×2048 (1:1). See supported resolution buckets for other aspect ratios.
For high-quality infographic generation, it is recommended to apply prompt enhancement before generating images.
python examples/editing/inference.py --model_path sensenova/SenseNova-U1-8B-MoT --prompt "Change the animal's fur color to a darker shade." --image examples/editing/data/images/1.webp --cfg_scale 4.0 --img_cfg_scale 1.0 --cfg_norm none --timestep_shift 3.0 --num_steps 50 --output output_edited.png --profile --compare💡 Pre-resize inputs to ~2048×2048 resolution with orginal aspect ratio before inference for best quality (seeexamples/editing/resize_inputs.py).
python examples/interleave/inference.py --model_path SenseNova/SenseNova-U1-8B-MoT --prompt "I want to learn how to cook tomato and egg stir-fry. Please give me a beginner-friendly illustrated tutorial." --resolution "16:9" --output_dir outputs/interleave/ --stem demo --profileSeeexamples/README.mdfor batched inference, JSONL format, prompt enhancement, resolution buckets, and full flag reference.
Seedocs/gpu_mem_profiler.mdfor GPU memory profiler.
transformers path. They can be combined freely.--gguf_checkpoint to any of the four inference scripts (t2i, editing, interleave, vqa) to load a quantized .gguf file via the diffusers GGUF Linear layer instead of the bf16 safetensors weights. The base --model_path is still required (for tokenizer / config / non-LM weights).1# install the optional extra once
2uv pip install -e ".[gguf]" # or: pip install "gguf>=0.10.0" "diffusers>=0.30.0"
3
4python examples/t2i/inference.py \
5 --model_path sensenova/SenseNova-U1-8B-MoT \
6 --gguf_checkpoint /path/to/SenseNova-U1-8B-MoT-Merger-Q4_K_M.gguf \
7 --prompt "A male peacock trying to attract a female" \
8 --output output.pngSenseNova-U1-8B-MoT-Merger (multiple quant levels: Q3 / Q4 / Q5 / Q6 / Q8) are available at:| Quantized weights | HF link |
|---|---|
| SenseNova-U1-8B-MoT-Merger-gguf | 🤗 smthem/SenseNova-U1-8B-MoT-Merger-gguf |
🙏 Thanks to GitHub user @smthem for contributing the quantized GGUF weights to the community.
--vram_mode: single-GPU layer offload--vram_mode to keep the language-model layers resident on CPU pinned memory and stream them onto the GPU on-demand during forward, freeing weight VRAM while keeping activations on-device.| Mode | Behavior | When to use |
|---|---|---|
full (default) | No offload; whole model on GPU | Plenty of VRAM, best speed |
low | Synchronous per-layer CPU↔GPU swap | Lowest VRAM footprint |
balanced | Async prefetch overlaps H2D copy with compute | Tight on VRAM but want to recover speed |
1python examples/t2i/inference.py \
2 --model_path sensenova/SenseNova-U1-8B-MoT \
3 --vram_mode balanced \
4 --prompt "..." --output output.png--gguf_checkpoint and --vram_mode compose: a Q4 GGUF + balanced is the recommended setup for ~10–12 GB consumer cards.TP2 + CFG2, this stack delivers roughly ~0.15 s/step and ~9 s end-to-end for a 2048×2048 image on H100 / H200, with a ~2.4–3.2× prefill speedup from our FA3-based hybrid-mask attention over the Triton baseline. Full per-GPU performance are reported in docs/inference_infra.md.docker pull lightx2v/lightllm_lightx2v:20260407⚙️ Deployment guide (Docker, launch flags, modes, quantization, API test): seedocs/deployment.md.📖 Full design and performance profiling: seedocs/inference_infra.md.
| Discord | WeChat Group |
 |  |