The HY 3D 2.1 workflow is designed to transform single 2D images into detailed 3D models using the Hunyuan3D 2.1 model by Tencent. This workflow leverages a series of specialized nodes to process an image and convert it into a 3D mesh. The process begins with the 'ImageOnlyCheckpointLoader' and 'LoadImage' nodes, which prepare the image for 3D transformation. The 'ModelSamplingAuraFlow' node then applies the Hunyuan3D model to generate a latent representation of the image in 3D space. This latent data is further processed by the 'EmptyLatentHunyuan3Dv2' and 'Hunyuan3Dv2Conditioning' nodes to refine the 3D structure.
The workflow continues with the 'KSampler' node, which samples the latent data to ensure the 3D model is as detailed and accurate as possible. The 'VAEDecodeHunyuan3D' node decodes this sampled data into a voxel representation, which is then converted into a mesh using the 'VoxelToMesh' node. Finally, the 'SaveGLB' node exports the 3D model in GLB format, ready for use in various applications. This workflow is particularly useful for artists and developers looking to create 3D content from existing 2D assets, offering a streamlined process that minimizes manual modeling efforts.
API
通过代码使用此工作流
每个 Comfy 工作流都是一个 JSON 图。下面的内容就是此工作流的完整数据，ComfyUI 运行时也是如此——你可以从该 URL 获取、放入版本控制，或在 ComfyUI 中加载并逐节点运行。
使用 Comfy SDK 从 TypeScript 或 Python 运行。相同代码可用于 Comfy Cloud 或你自托管的 ComfyUI——只需更改基础 URL。
// Install (beta)
npm i @comfyorg/sdk
// Run "混元3D 2.1" (TypeScript)
import { Comfy } from "@comfyorg/sdk";
const client = new Comfy({ apiKey: "comfyui-..." });
// This workflow, exported in API format (see note below)
const wf = await client.workflows.fromFile("3d_hunyuan3d-v2.1_api.json");
const asset = client.assets.fromFile("input.png");
wf.setInput("2", "image", asset); // LoadImage
const job = await client.run(wf);
await job.getOutputs("10")[0].toFile("output.png"); // SaveGLBSDK 需要 API 格式的工作流：在 ComfyUI 中打开此工作流，使用 文件 → 导出工作流（API）。
常见问题
