[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-velum-cxx-cosyvoice3-tts":3,"topics-all":35,"news-related-b3ce6178-954b-403d-b4c3-6480888a996d":64},{"id":4,"title":5,"summary":6,"content":7,"original_url":8,"source_id":9,"tags":10,"translations":21,"news_slug":28,"published_at":29,"created_at":30,"modified_at":31,"is_published":32,"publish_type":33,"image_url":14,"view_count":34},"b3ce6178-954b-403d-b4c3-6480888a996d","Velum 把 CosyVoice3 重写成单文件 C++ 推理程序","HardenedLinux 开源 Velum,把 CosyVoice3 TTS 用 modern C++ 重写,神经网络跑在 GGML+CUDA,运行时不再依赖 Python。四阶段逐级对齐 PyTorch 参考到 GREEN(相对误差 ≤1%),端到端在 CPU 与 CUDA 上均通过校验。","CosyVoice3 是当下表现靠前的开源 TTS 模型,但它的推理栈由 Python 生态堆叠,部署时要把 PyTorch、依赖包、版本冲突一起装好,十几 GB 的运行时是常态。HardenedLinux 在 GitHub 开源的 Velum 直接把这套流水线用 modern C++ 重写,产物只有一个可执行文件,运行时不再依赖 Python。\n\n## 工程取舍:在线零 Python\n\n神经网络前向基于 GGML+CUDA,推理阶段不调用任何 Python,只有数据预处理阶段(从文本里抽出 token、构造 mel-spectrogram 候选)继续用 Python 跑一次,生成离线资产。运行时目录里不再需要装 PyTorch、依赖包、版本冲突解释器,所谓\"部署更新\"就是拷贝这一个可执行文件。\n\n## 四阶段对齐 PyTorch 参考\n\nVelum 设计了一个四阶段递进式对齐流程,逐级把推理误差压到 PyTorch 参考实现的 GREEN 阈值(相对误差 ≤1%)。源材料显示,这套流水线在 CPU 和 CUDA 上均通过所有 GREEN 检查,没有依赖任何 GPU 算子替换或黑盒降级;实现细节全部以源码形式公开,可在 GitHub 仓库直接验证。\n\n## 谁是 Velum 的\"架构师\"?\n\nVelum 不只是给 CosyVoice3 写一个 C++ 包装。Solidot 引用的报道指出,它的整体架构由人类做规划、再交给 DeepSeek-v4-pro 完成。这意味着 Velum 既是一份高质量的 TTS 推理程序,也是一次 LLM 主导大型工程项目的案例:人类定调、模型填血肉,最后产出一份可在生产环境部署的可执行文件。\n\n## 一个边缘部署的解法\n\nCosyVoice3 是开源 TTS 的旗舰级别模型,但它的 Python 推理栈让 Agent 集群、边缘设备、自托管小团队的部署成本过高。Velum 通过 C++ + GGML 的组合把整套推理栈压成一个可执行文件,实际意义是:Agent 想要集成语音能力时,不再需要拖一个十几 GB 的 Python 运行时,只要把这个二进制复制过去即可。","https:\u002F\u002Fgithub.com\u002Fhardenedlinux\u002Fvelum","998df6db-96e6-4b8e-8be1-cfa00a6cd177",[11,15,18],{"id":12,"name":13,"slug":13,"description":14,"color":14},"7ac06d8e-b074-4147-abfc-ffaa4c6b8744","ai-efficiency",null,{"id":16,"name":17,"slug":17,"description":14,"color":14},"fca9258a-9430-455a-b95d-b9fae5e373a8","ai-inference",{"id":19,"name":20,"slug":20,"description":14,"color":14},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",[22],{"id":23,"lang":24,"title":25,"summary":26,"content":27},"ded5fa56-f27c-4762-a8f4-30b5e5ab225d","en","Velum: native C++ runtime for CosyVoice3, zero Python","HardenedLinux open-sources Velum: a modern C++ reimplementation of the CosyVoice3 TTS pipeline. Neural networks run on GGML+CUDA, the runtime needs no Python interpreter. A four-stage numerical verification brings every stage within GREEN (≤1% relative error) against the PyTorch reference on both CPU and CUDA.","CosyVoice3 is one of the strongest open-source TTS models today, but its inference stack is glued to the Python ecosystem: PyTorch, dependency packages, and version conflicts travel together, and a multi-gigabyte runtime is the norm for deployment. Velum, open-sourced on GitHub by HardenedLinux, rewrites the entire pipeline in modern C++ and ships a single executable whose runtime needs no Python.\n\n## Engineering trade-off: zero Python at runtime\n\nNeural-network forward passes run on GGML with CUDA acceleration; the inference path never calls Python. Only the offline preprocessing stage (token extraction and mel-spectrogram candidate construction) is still done in Python, producing frozen asset files the C++ binary loads verbatim. The runtime directory no longer needs PyTorch, dependency packages, or version-pin workarounds — \"deploy an update\" means copy the binary.\n\n## Four-stage alignment with the PyTorch reference\n\nVelum ships a four-stage progressive verification flow that pushes the inference error against the PyTorch reference down to the GREEN threshold (relative error ≤ 1%). The source materials show the pipeline passes every GREEN check on both CPU and CUDA, without resorting to opaque GPU-operator substitutions or black-box downgrades. Implementation details are published as source and can be audited directly in the GitHub repository.\n\n## Who is Velum's architect?\n\nVelum is more than a C++ wrapper around CosyVoice3. The Solidot write-up notes that the overall architecture was planned by a human, then handed off to DeepSeek-v4-pro for implementation. That makes Velum both a high-quality TTS inference binary and a case study in LLM-driven large-project engineering: a human sets the direction, the model fills in the muscle, and the result is a production-ready executable.\n\n## An answer for edge deployment\n\nCosyVoice3 sits at the top of the open-source TTS stack, but its Python runtime makes it expensive to drop into agent fleets, edge devices, or small self-hosted teams. Velum compresses the entire inference stack into a single executable via C++ + GGML. The practical consequence: when an agent wants to add voice capability, it no longer has to drag a multi-gigabyte Python runtime along — just copy the binary over.","velum-cxx-cosyvoice3-tts","2026-09-26T07:00:00Z","2026-09-26T11:06:51.765162Z","2026-09-26T11:06:51.765170Z",true,"agent",459,[36,46,55],{"slug":37,"tag_slug":38,"title_zh":39,"title_en":40,"intro_zh":41,"intro_en":42,"id":43,"is_active":32,"created_at":44,"modified_at":45},"image-generation","text-to-image","图像生成：从文生图到逐框编辑的模型竞赛","Image Generation: from text-to-image to frame-level editing","文生图已是红海，战场正转向编辑与可控性：FLUX.2\u002F3 用开源和量化把生图搬进消费级显卡，Qwen-Image、Seedream 等国产模型紧咬不放，GPT Image 2 则把推理能力带进像素。本专题追踪图像生成与编辑模型的发布、架构演进、加速量化与商业化数据，看「好图」如何变成「可用的图」。","Text-to-image is a crowded race, and the frontier has moved to editing and control: FLUX.2\u002F3 push open weights and quantization onto consumer GPUs, Qwen-Image and Seedream close in from China, and GPT Image 2 brings reasoning into pixels. This topic tracks model releases, architecture shifts, acceleration work, and the business numbers behind \"pretty\" becoming \"usable\".","52de7c59-7864-4569-8ae7-3d656d26a1a1","2026-11-20T07:20:23.534506Z","2026-11-20T07:20:23.534527Z",{"slug":47,"tag_slug":47,"title_zh":48,"title_en":49,"intro_zh":50,"intro_en":51,"id":52,"is_active":32,"created_at":53,"modified_at":54},"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":56,"tag_slug":56,"title_zh":57,"title_en":58,"intro_zh":59,"intro_en":60,"id":61,"is_active":32,"created_at":62,"modified_at":63},"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":65},[66,71,76,81,86,91],{"id":67,"title":68,"news_slug":69,"published_at":70},"ab1e6bdf-9433-41f0-be4d-cc0f78523ac2","SonicBoom v0.0.3:一个 .so 干掉手工调优","sonicboom-v003-mlir-compiler","2026-10-19T09:03:05+00:00",{"id":72,"title":73,"news_slug":74,"published_at":75},"bd0a85ff-0a8c-4900-a8fc-65e606a8bce0","Velum:零 Python 跑 CosyVoice3","velum-cosyvoice3-cpp-zero-python","2026-09-26T07:00:00+00:00",{"id":77,"title":78,"news_slug":79,"published_at":80},"6fa1bc74-c98e-476b-bc4c-9ae057105ffb","ParaTempo:免训练并行推理,延迟最高降 32%、token 省三成","paratempo-temporal-confidence-parallel-reasoning","2026-08-24T17:20:00+00:00",{"id":82,"title":83,"news_slug":84,"published_at":85},"761043fc-3329-42e4-914d-3f567b665b90","面壁智能联合 OpenBMB 开源 ForgeStencil：用双 Agent 把 Stencil 优化直接打进真实科学计算软件","mianbiai-openbmb-forgestencil-stencil-optimization","2026-08-04T10:00:00+00:00",{"id":87,"title":88,"news_slug":89,"published_at":90},"f9a6c57b-a98e-407c-9751-93907ff314b0","Capacity-Aware Inference：MoE 推理 1.85× 加速","iclr-2026-capacity-aware-moe","2026-07-27T10:00:00+00:00",{"id":92,"title":93,"news_slug":94,"published_at":95},"ba729d8d-d18e-4e77-824e-1db99ec469eb","近零词嵌入暗坑:五成门控检查点训练静默白跑","near-zero-embeddings-gated-training-stall","2026-11-04T15:20:36+00:00"]