Hunyuan 3D v3.1 provides Pro and Rapid text/image workflows for a textured object or a geometry-only study. Describe the object in words or provide one primary picture, then choose the generation type and PBR materials for the intended result. Pro adds a target face count and, in its image task, seven named optional views to clarify the object from additional directions. Rapid provides its own text and image tasks with a geometry switch. Compare the accepted controls and complete estimate, then inspect shape and materials from several angles.
Hunyuan 3D v3.1 is developed by Tencent. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.Select text generation for a written object concept or image generation for exactly one primary source. Choose Pro when the request needs its face-count or named-view controls; choose Rapid for its corresponding text/image and geometry options. Both variants support PBR in textured generation, with their own base amount and applicable additions visible in the task estimate.
Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials.
Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials.
Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials.
Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials.
Text prompts define form, parts and materials, with Pro accepting up to 1,024 characters and Rapid up to 200. Image tasks require one image_urls entry. Pro Image to 3D can add back, left, right, top, bottom, left-front and right-front image URLs, keeping the object’s design and proportions consistent across views. Choose the input method and geometry/material settings before inspecting the full request cost.
| Model / task | Input and workflow |
|---|---|
| Hunyuan 3D v3.1 Pro Text to 3D | Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials. |
| Hunyuan 3D v3.1 Pro Image to 3D | Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials. |
| Hunyuan 3D v3.1 Rapid Text to 3D | Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials. |
| Hunyuan 3D v3.1 Rapid Image to 3D | Generate 3D assets from text or images with Pro face-count and multi-view controls, Rapid workflows, and PBR materials. |
Existing records include a bronze censer, walnut accordion, geometry-only sandstone arch and millstone, plus a railway signal. The preview images come from those original generations. The two full cases provide actual returned models for inspecting the text-based censer and image-guided accordion, each with its own inputs and settings.
Use a prompt to specify connected form, relief, wear and material regions, then select Normal and PBR when textured materials are intended. The censer record requests bronze patina, guardian relief and loop handles; inspect the actual model and its associated material files.
Provide exactly one primary image and choose Pro generation type and face count. Add named views when the request needs more visual evidence. The accordion record uses its actual front reference with the default Pro face count, giving you visible parts and materials to compare with the model.
Choose generate_type Geometry for a geometry-only object and set face_count from 40,000 to 1,500,000. The sandstone-arch record uses 40,000 faces to request connected pillars and an open passage, so you can assess silhouette and openings as a shape study.
Prepare a text prompt up to 200 characters and enable_geometry true for a geometry-only study. The millstone record describes a square opening and radial grooves; inspect the preview and returned model for the requested solid form and surface structure.
Provide one image and use enable_geometry false for textured generation, adding PBR when needed. The signal record supplies an octagonal railway object; compare the recorded preview with its source to review the silhouette and visible material regions.
The censer request supplies a complete bronze-artifact prompt with Normal generation, the default Pro face count and PBR. The accordion request supplies one actual source image with Normal generation and its recorded face/material settings. Read the full prompt or original input object, compare the source with the preview and rotate the actual returned GLB to inspect form, surface and the back of each independent object.
A freestanding ancient bronze incense censer, squat rounded vessel on three stout feet, two thick upright loop handles and a fitted domed lid. Deep symmetrical guardian-beast relief around the body, raised spiral bands, aged green patina in recesses and worn golden bronze on edges. One complete museum artifact, no smoke, no text, no pedestal.
{ "generate_type": "Normal", "face_count": 500000, "enable_pbr": false, "image_urls": [ "https://cdn.vidgo.ai/apis/models/hunyuan-3d/v3.1/pro/image-to-3d/v1/02/input.png" ] }
Prepare a written object or a clear primary image, choose Pro or Rapid and configure textured or geometry-only generation. Inspect the returned asset and related files for artifact studies, architectural miniatures, stylized props, geological forms, mechanical objects and scene-development concepts. Use the Pro face and view controls when they support the project.
Describe the object’s connected form, relief and aged material regions, then choose textured generation and PBR when appropriate. Inspect geometry and materials together for a museum or visual-research concept.
Prepare a compact building or structural object with clear openings and connected parts. Use text or a primary image, then inspect silhouette and surfaces in the generated model for a scene proposal.
Describe a self-contained fantasy or handcrafted prop and its major parts. Select textured or geometry-only output and review the result from multiple angles before refining its design.
Write the rock structure, openings, grooves and stable supports, then choose geometry generation. Inspect the connected form and face-count choice for a geological specimen or environment concept.
Provide a clear primary image and optional Pro named views of the same design. Review the resulting parts, thickness and material regions against all sources for a mechanical-object study.
Match the object brief and generation controls to the planned scene. Retrieve the model and associated materials, inspect them in the intended 3D application and use the observations to plan the next asset revision.
Pro and Rapid use their own current per-generation base amounts, shown in the dynamic table. The complete estimate adds PBR for textured generation, a Pro face-count addition when the target differs from the 500,000 default, and one Pro image-view addition when any named extra view is supplied. Geometry-only generation uses its shape settings; the PBR addition applies to textured generation. Open the exact task with all inputs to review the full amount before submitting.
| Model | Provider | Workflow | Input material | Output contents | Starting rate |
|---|---|---|---|---|---|
| Hunyuan 3D v3.1 Pro Text to 3D | Tencent | Text to 3D | Object prompt | 3D model / material files | $0.3 / generation |
| Hunyuan 3D v3.1 Pro Image to 3D | Tencent | Image to 3D | 1 primary; up to 7 named views | 3D model / material files | $0.3 / generation |
| Hunyuan 3D v3.1 Rapid Text to 3D | Tencent | Text to 3D | Object prompt | 3D model / material files | $0.18 / generation |
| Hunyuan 3D v3.1 Rapid Image to 3D | Tencent | Image to 3D | Exactly 1 primary image | 3D model / material files | $0.18 / generation |
Test your creative brief online, then bring the verified request into your application.
Open the selected task and enter the prompt or required primary image with any supported Pro views. Configure the generation type, materials and available face control, then review the full amount. Generate and inspect the preview and model from several angles, retrieving all associated files for shape and material checks in the target workflow.
Choose a task, inspect its settings and cost, and follow the result from an online test to an authenticated API request.
Choose Pro or Rapid, then text or one primary image. Pro can select Normal or Geometry and face count; its image task can add named views. Rapid uses enable_geometry for textured or geometry-only output. Enable PBR when requesting textured materials, then review the complete estimate including applicable additions.
Open the selected task and enter the prompt or required primary image with any supported Pro views. Configure the generation type, materials and available face control, then review the full amount. Generate and inspect the preview and model from several angles, retrieving all associated files for shape and material checks in the target workflow.
Start from the selected variant’s current base amount and include textured PBR, a nondefault Pro face count and any Pro image-view addition when applicable. Use the task estimate with the complete inputs to budget for each object variation.
Read the actual prompt or recorded input parameters and compare source images with the model preview and interactive result. Inspect silhouette, geometry and materials from several angles before adapting the workflow to your object.
Retain the task ID returned by the API to query status and retrieve completed file URLs. Use your online generation history to review earlier results and the creative direction you tested.
Use the documentation linked from each task for its exact model identifier, accepted inputs, authentication and response format. Integrate the same task you tested online so its media and settings match the request.
Hunyuan 3D v3.1 is a Tencent 3D family on Vidgo with Pro and Rapid text/image tasks. Both provide textured and geometry-only workflows with PBR for textured materials. Pro adds face-count control and optional named image views, using its own base and addition rules.
Read 3D-model introductions, API guides and platform comparisons to understand source preparation, mesh and material controls, integration and generation costs.
