国产芯片训练的超轻量文生图模型，手机本地运行仅需1.6秒。
JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators
- 用0.387亿参数小模型+4步采样，实现手机端快速生成。
- 在中文提示下直接生成，无需翻译，性能超越多个大模型。
- 全程国产算力训练，支持离线部署，适合隐私敏感场景。
文生图扩散模型通常依赖大量计算资源和云端基础设施，难以在边缘设备上高效部署。本文提出JuZhou 1.0，首个完全基于中国自主研发算力训练的轻量级文生图基础模型，支持全离线、本地化执行。其核心设计包括：（1）紧凑的图像生成主干网络，包含0.385亿参数的去噪U-Net与190万参数的蒸馏解码器，总参数约0.387亿；（2）采用修正流训练结合DMD2蒸馏，将推理步骤减少至4步；（3）基于900万条精选图文对训练中文语义对齐，支持直接中文提示生成；（4）整个训练与蒸馏流程在国产Sugon K100 AI加速器上完成，无需依赖NVIDIA GPU。尽管规模极小，其28步基线模型在GenEval评测中得分为0.69，优于SDXL（2.6B, 0.55）、SD3-Medium（2B, 0.62）和IF-XL（4.3B, 0.61）。我们进一步在Android端验证了完整诗歌到图像流程，在搭载骁龙8 Elite Gen 5平台的手机上，4步去噪分支耗时约1.6秒，完整诗歌生成流程（含本地提示优化）在小米17 Pro Max上为4.5秒。该工作为中文原生生成、国产算力训练及完全离线部署提供了可行方案。
原文摘要 · Abstract (English)
Text-to-image (T2I) diffusion models typically require substantial computational resources and cloud infrastructure, posing significant challenges for edge deployment in terms of latency, cost, and user privacy. We present JuZhou 1.0, an ultra-lightweight T2I foundation model designed for fully offline, on-device execution. JuZhou 1.0 achieves its efficiency through four key designs: (1) a compact image-generation backbone consisting of a 0.385B-parameter denoising U-Net and a 1.90M-parameter distilled decoder, totaling approximately 0.387B parameters; (2) Rectified Flow training combined with DMD2 distillation, reducing inference to 4 sampling steps; (3) Chinese semantic alignment trained on 9M curated image-text pairs, enabling direct Chinese prompting without external translation at inference time; and (4) a training and distillation pipeline completed on domestically developed Sugon K100 AI accelerators without relying on NVIDIA GPUs for training or distillation. Despite its compact scale, the 28-step base model of JuZhou 1.0 achieves an overall GenEval score of 0.69, outperforming published baselines including SDXL (2.6B, 0.55), SD3-Medium (2B, 0.62), and IF-XL (4.3B, 0.61). We further validate the full poetry-to-image pipeline on Android and the core CLIP-U-Net-VAE generation branch on iOS. On a smartphone powered by the Snapdragon 8 Elite Gen 5 Mobile Platform, the 4-step U-Net denoising branch runs in approximately 1.6 seconds, while the full Android poetry-to-image pipeline takes 4.5 seconds with on-device prompt refinement on Xiaomi 17 Pro Max. These results position JuZhou 1.0 as a practical approach to mobile text-to-image generation and provide a concrete reference for Chinese-native generation, domestic-compute training, and fully offline on-device deployment after one-time installation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。