Buckets:
| import"../chunks/DsnmJJEf.js";import{i as v,h as b,C as _,H as o,E as w,s as A}from"../chunks/DdZvggmf.js";import{p as E,o as V,s as t,f as x,a as p,b as y,c as d,n as T}from"../chunks/BbekZcyp.js";const I='{"title":"混合推理","local":"混合推理","sections":[{"title":"为什么使用混合推理?","local":"为什么使用混合推理","sections":[],"depth":2},{"title":"可用模型","local":"可用模型","sections":[],"depth":2},{"title":"集成","local":"集成","sections":[],"depth":2},{"title":"更新日志","local":"更新日志","sections":[],"depth":2},{"title":"内容","local":"内容","sections":[],"depth":2}],"depth":1}';var C=d('<meta name="hf:doc:metadata"/>'),U=d(`<p></p> <!> <!> <p><strong>通过混合推理赋能本地 AI 构建者</strong></p> <blockquote class="tip"><p>混合推理是一项<a href="https://huggingface.co/blog/remote_vae" rel="nofollow">实验性功能</a>。 | |
| 可以在此处提供反馈<a href="https://github.com/huggingface/diffusers/issues/new?template=remote-vae-pilot-feedback.yml" rel="nofollow">此处</a>。</p></blockquote> <!> <p>混合推理提供了一种快速简单的方式来卸载本地生成需求。</p> <ul><li>🚀 <strong>降低要求:</strong> 无需昂贵硬件即可访问强大模型。</li> <li>💎 <strong>无妥协:</strong> 在不牺牲性能的情况下实现最高质量。</li> <li>💰 <strong>成本效益高:</strong> 它是免费的!🤑</li> <li>🎯 <strong>多样化用例:</strong> 与 Diffusers � 和更广泛的社区完全兼容。</li> <li>🔧 <strong>开发者友好:</strong> 简单请求,快速响应。</li></ul> <hr/> <!> <ul><li><strong>VAE 解码 🖼️:</strong> 快速将潜在表示解码为高质量图像,不影响性能或工作流速度。</li> <li><strong>VAE 编码 🔢:</strong> 高效将图像编码为潜在表示,用于生成和训练。</li> <li><strong>文本编码器 📃(即将推出):</strong> 快速准确地计算提示的文本嵌入,确保流畅高质量的工作流。</li></ul> <hr/> <!> <ul><li><strong><a href="https://github.com/vladmandic/sdnext" rel="nofollow">SD.Next</a>:</strong> 一体化 UI,直接支持混合推理。</li> <li><strong><a href="https://github.com/kijai/ComfyUI-HFRemoteVae" rel="nofollow">ComfyUI-HFRemoteVae</a>:</strong> 用于混合推理的 ComfyUI 节点。</li></ul> <!> <ul><li>2025 年 3 月 10 日:添加了 VAE 编码</li> <li>2025 年 3 月 2 日:初始发布,包含 VAE 解码</li></ul> <!> <p>文档分为三个部分:</p> <ul><li><strong>VAE 解码</strong> 学习如何使用混合推理进行 VAE 解码的基础知识。</li> <li><strong>VAE 编码</strong> 学习如何使用混合推理进行 VAE 编码的基础知识。</li> <li><strong>API 参考</strong> 深入了解任务特定设置和参数。</li></ul> <!> <p></p>`,1);function q(f,m){E(m,!1),V(()=>{new URLSearchParams(window.location.search).get("fw")}),v();var e=U();b("tvqezj",h=>{var c=C();A(c,"content",I),p(h,c)});var l=t(x(e),2);_(l,{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"});var a=t(l,2);o(a,{title:"混合推理",local:"混合推理",headingTag:"h1"});var i=t(a,6);o(i,{title:"为什么使用混合推理?",local:"为什么使用混合推理",headingTag:"h2"});var n=t(i,8);o(n,{title:"可用模型",local:"可用模型",headingTag:"h2"});var s=t(n,6);o(s,{title:"集成",local:"集成",headingTag:"h2"});var r=t(s,4);o(r,{title:"更新日志",local:"更新日志",headingTag:"h2"});var g=t(r,4);o(g,{title:"内容",local:"内容",headingTag:"h2"});var u=t(g,6);w(u,{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/zh/hybrid_inference/overview.md"}),T(2),p(f,e),y()}export{q as component}; | |
Xet Storage Details
- Size:
- 3.78 kB
- Xet hash:
- 2dad28ac9198025618618287bc72796e0424b3100afce0fc89da56f4951e8036
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.