[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-moss-image-bilingual-text-rendering-base":3,"topics-all":38,"news-related-3bf416bb-a989-4b02-8341-dbafc7c96c48":57},{"id":4,"title":5,"summary":6,"content":7,"original_url":8,"source_id":9,"tags":10,"translations":24,"news_slug":31,"published_at":32,"created_at":33,"modified_at":34,"is_published":35,"publish_type":36,"image_url":14,"view_count":37},"3bf416bb-a989-4b02-8341-dbafc7c96c48","MOSS-Image 开源:5B 双语生图,文字排版可编程","OpenMOSS 放出 5B 生图基座 MOSS-Image-Base 与 SFT 版,Apache 2.0。模型原生支持中英双语提示词与 4K token 长输入,主打图内文字渲染,可用 JSON layout 精确控制文字与物体位置;官方未公布 benchmark,衍生微调已现。","10 月 16 日,OpenMOSS 团队在 GitHub 官宣 MOSS-Image 的 Base 与 SFT 两版,权重同步上架 Hugging Face,权重与推理代码均按 Apache 2.0 放开。在 FLUX、Qwen-Image 新品不断的开源生图赛道,这个 5B 参数的 DiT 基座走差异化路线:不拼通用美感,主攻图内文字渲染与排版控制,原生支持中英双语提示词。\n\n## 一个 5B 的双语基座\n\nMOSS-Image-Base 是 MOSS-Image 家族的基座模型,官方定位是给开发者和研究者做下游微调的底座,而不是开箱即用的消费级产品。核心规格:5B 参数 DiT 主干,transformer 与文本编码器 bfloat16、VAE float32,捆绑 VAE 直接用 FLUX.2 small decoder——README 致谢里点名 Diffusers、Qwen 和 FLUX 三团队,技术血统相当坦诚。\n\n提示词侧有两个实用设计:一是原生中英双语,官方 quick start 示例就是一个中文手写便签——冰箱贴上写着「记得买牛奶和鸡蛋」;二是官方称支持最长 4K token 提示词,tag 序列也能直接喂。\n\n## JSON layout:把排版变成可编程输入\n\n真正有辨识度的是 layout 控制。MOSS-Image 接受一种 JSON 结构作为 prompt:`summary.caption` 描述整体场景,`items` 数组逐个声明元素,每个元素带 `label`、`caption` 和 `bbox`,需要渲染的文字放在 `text` 字段里。坐标系用 `yxyx` 格式、以画布中心为原点的绝对像素坐标——1024×1024 画布上每轴从 -512 到 +512。也就是说,「这行字写在便利贴上、便利贴贴在冰箱左上角」这种需求,从自然语言猜测变成了可精确编辑的 JSON:改文字、改描述、改框位置都是结构化操作,同一 layout 还能做受控编辑。\n\n采样配置官方也给全了:flow-matching Euler,50 步,CFG 3,shift 5,负分支默认跳过第 22 层(0 起数),出图尺寸须为 16 的倍数;依赖只要 diffusers 0.38+、transformers 5.3+,几行代码就能跑。\n\n## 生态起量很快,benchmark 还缺席\n\n上线至今,HF 页面已挂 16 个微调衍生权重、9 个 adapter、2 个量化版、12 个 Spaces 演示,GitHub 仓库 485 star;ComfyUI 原生插件已进 ComfyUI Manager,合入主仓库也在推进。仓库还附一个「prompt enhancer」agent skill,为 GPT-6 家族 agent 设计,把模糊想法编译成详细 caption 或 layout JSON。官方亮点里还明确写了「异构 GPU 训练」——训练在替代 GPU 平台上完成,在当前算力格局下是个不小的信号。\n\n冷水也明确:OpenMOSS 未随发布公布 benchmark 与支持分辨率清单,参数之外的横向对比暂无从做起;「强文字渲染」目前是官方自报,第三方复现数据还没出现。\n\n## 所以呢\n\nMOSS-Image 的价值不在「又开源一个 5B 生图模型」,而在把「图内排版」做成结构化输入——海报、电商详情页、带字表情包这类中文场景刚需,是不少西方开源模型没认真覆盖的空白。Base 加 SFT 双版本、Apache 2.0、微调门槛低,这是一套标准的「社区帮我长出生态」打法。等第三方跑分和 LoRA 实测出来,这个 5B 的真实水位才见分晓;但双语文字渲染这个切口,已经足够它在不拥挤的位置站住了。","https:\u002F\u002Fgithub.com\u002FOpenMOSS\u002FMOSS-Image","ef16cd32-fd57-41c2-a516-9c81b3835576",[11,15,18,21],{"id":12,"name":13,"slug":13,"description":14,"color":14},"7b67033c-19e6-4052-a626-e681bba64c7a","diffusion",null,{"id":16,"name":17,"slug":17,"description":14,"color":14},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",{"id":19,"name":20,"slug":20,"description":14,"color":14},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",{"id":22,"name":23,"slug":23,"description":14,"color":14},"c883fd20-1d66-4fb7-9fc7-320fa7f87023","text-to-image",[25],{"id":26,"lang":27,"title":28,"summary":29,"content":30},"3487d92d-fbd2-4a04-a74b-2b8d58e2928d","en","MOSS-Image Open-Sourced: 5B Bilingual Base with JSON Layout","OpenMOSS ships a 5B Apache-2.0 DiT image base: bilingual prompts, 4K-token inputs, in-image text rendering, JSON layout control; no benchmarks yet.","On October 16, the OpenMOSS team announced Base and SFT versions of MOSS-Image on GitHub, with weights simultaneously published on Hugging Face, both weights and inference code under Apache 2.0. In an open image-generation field crowded with new FLUX and Qwen-Image releases, this 5B-parameter DiT base takes a differentiated route: instead of chasing general aesthetics, it focuses on in-image text rendering and layout control, with native bilingual prompt support.\n\n## A 5B bilingual base\n\nMOSS-Image-Base is the base model of the MOSS-Image family, positioned by the team as a foundation for developers and researchers to fine-tune downstream rather than a finished consumer product. Core specs: a 5B-parameter DiT backbone, bfloat16 for the transformer and text encoder, float32 VAE, and a bundled VAE that directly reuses the FLUX.2 small decoder — the README acknowledgements name the Diffusers, Qwen, and FLUX teams, a fairly honest disclosure of its technical lineage.\n\nThe prompt side has two practical designs. First, native English and Chinese support: the official quick-start example is a Chinese handwritten note — a fridge sticky note reading \"remember to buy milk and eggs\". Second, the team says prompts up to 4K tokens are accepted, and tag sequences can be fed directly.\n\n## JSON layout: typesetting as a programmable input\n\nThe genuinely distinctive part is layout control. MOSS-Image accepts a JSON structure as the prompt: `summary.caption` describes the overall scene, the `items` array declares elements one by one, each with `label`, `caption`, and `bbox`, and any text to render goes in the `text` field. Coordinates use `yxyx` format with absolute pixel coordinates centered on the canvas — on a 1024×1024 canvas each axis runs from -512 to +512. A request like \"this line of text on a sticky note, the note on the fridge's upper-left corner\" thus turns from natural-language guesswork into precisely editable JSON: change the text, the description, or the box position, and the composition adjusts in a structured way; controlled editing on the same layout is also supported.\n\nThe official sampling configuration is fully disclosed: flow-matching Euler, 50 steps, CFG 3, shift 5, the negative branch skips layer 22 (zero-based) by default, and output dimensions must be multiples of 16. Dependencies are just diffusers 0.38+ and transformers 5.3+; a local custom pipeline runs in a few lines.\n\n## Fast ecosystem growth, absent benchmarks\n\nSince launch, the Hugging Face page already lists 16 fine-tuned derivatives, 9 adapters, 2 quantized versions, and 12 Spaces demos, while the GitHub repo sits at 485 stars. A native ComfyUI plugin is available in ComfyUI Manager, and integration into the official ComfyUI repository is in progress. The repo also ships a \"prompt enhancer\" agent skill, designed for GPT-6-family agents, that compiles vague image ideas into detailed captions or layout JSON. Another detail worth noting: the official highlights explicitly list heterogeneous GPU training — the model was trained on alternative GPU platforms, a non-trivial signal under the current compute landscape.\n\nThe cold water is equally clear. OpenMOSS did not publish benchmark results or a supported-resolution list alongside the release, so horizontal comparison beyond parameter count is not yet possible; \"strong text rendering\" remains a self-reported claim until third-party reproduction data appears.\n\n## So what\n\nThe value here is not \"yet another open 5B image model\", but that MOSS-Image turns in-image typesetting into structured input — posters, e-commerce detail pages, and captioned memes, the everyday needs of Chinese-language scenarios, are exactly the gap that quite a few Western open models never seriously covered. Base plus SFT dual releases, Apache 2.0, and low fine-tuning barriers form a standard \"let the community grow the ecosystem\" playbook. The 5B model's true standing will only be clear once third-party benchmarks and LoRA tests arrive; but the bilingual text-rendering wedge is already enough for it to hold uncrowded ground.","moss-image-bilingual-text-rendering-base","2026-10-18T19:08:44Z","2026-10-18T19:08:57.779589Z","2026-10-18T19:08:57.779602Z",true,"agent",178,[39,48],{"slug":40,"tag_slug":40,"title_zh":41,"title_en":42,"intro_zh":43,"intro_en":44,"id":45,"is_active":35,"created_at":46,"modified_at":47},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":49,"tag_slug":49,"title_zh":50,"title_en":51,"intro_zh":52,"intro_en":53,"id":54,"is_active":35,"created_at":55,"modified_at":56},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":58},[59,64,69,74,79,84],{"id":60,"title":61,"news_slug":62,"published_at":63},"bbbda7ff-b1c7-4142-b37f-2b3fd0547e02","Fibo 2.0 开源:生成编辑合一,登顶开源编辑榜","bria-fibo-2-0-unified-editing-generation","2026-10-21T21:08:11+00:00",{"id":65,"title":66,"news_slug":67,"published_at":68},"63299a5b-f3b7-40c4-9985-0132e0cbffac","多轮图像编辑漂移的解法：MT Forcing 用自回归扩散把单轮质量带进多轮","mt-forcing-multi-turn-image-editing","2026-10-20T05:06:17+00:00",{"id":70,"title":71,"news_slug":72,"published_at":73},"b88d9e4c-1450-4349-a3cd-c2f8c0ce50eb","FLUX 3 Klein 开源:5.6B,8GB 显存能跑","flux-3-image-klein-open-weights","2026-10-17T13:20:00+00:00",{"id":75,"title":76,"news_slug":77,"published_at":78},"5a46f955-86f3-41b9-8e88-1b31e948469e","上海AI Lab开源统一视觉dLLM:激活参数仅1.4B","intern-lumina-u2-multi-codebook-diffusion","2026-10-16T17:09:52+00:00",{"id":80,"title":81,"news_slug":82,"published_at":83},"162c36d7-37de-4906-b0dd-cd5ec6f9f455","微软擦除基础模型LensErase:一图一掩码,万物皆可擦","lenserase-erase-anything-foundation-model","2026-10-15T21:08:38+00:00",{"id":85,"title":86,"news_slug":87,"published_at":88},"d5ba4b97-e08e-4ddb-80d8-70f8060f6d9a","AgentGarten:把世界写成代码,智能体4轮学会躲猫猫","agentgarten-code-worlds-for-evolving-agents","2026-10-10T17:10:00+00:00"]