Game-asset pipelines
Reduce dense generated or sculpted assets into cleaner meshes that are easier to carry into real-time workflows.
Independent tool overview
Hunyuan 3D Smart Topology (PolyGen) at a glance
Hunyuan 3D PolyGen is Tencent's AI retopology technology, now available through Hunyuan 3D Global's Smart Topology workflow and API using the Polygen 1.5 model.
Visit the official Hunyuan 3D Smart Topology (PolyGen) site ↗
- Current model
- Polygen 1.5
- Input formats
- GLB or OBJ by URL, up to 200 MB
- Topology options
- Triangles or mixed triangles and quads
- API cost
- 50 credits per Smart Topology job
- Last reviewed
- August 28, 2026
Overview
What Hunyuan 3D Smart Topology (PolyGen) is
Tencent introduced Hunyuan3D-PolyGen as an art-grade 3D model focused on a stubborn production problem: converting dense geometry into cleaner, lower-polygon topology. The current Tencent Cloud implementation exposes that capability as Smart Topology, powered by Polygen 1.5.
This is a post-processing tool, not a text-to-3D generator by itself. You submit an existing high-poly GLB or OBJ model, choose triangular or mixed triangle-and-quad topology and a reduction level, then retrieve the completed job.
The best fit is a studio or 3D pipeline that needs repeatable API-driven retopology. Individual artists who want visual editing, manual edge-flow control, or a broad creation suite may prefer a desktop modeling tool or a more creator-oriented hosted platform.
Use cases
Who Hunyuan 3D Smart Topology (PolyGen) is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
3D production teams
Automate repeatable retopology jobs through Tencent Cloud instead of processing every asset manually.
Hunyuan 3D users
Post-process high-poly assets generated elsewhere in the Hunyuan ecosystem with a directly related topology service.
Capabilities
Core Hunyuan 3D Smart Topology (PolyGen) features
AI smart topology
Transforms a non-topologized high-poly input into a neater, lower-polygon textured geometry model.
Triangle or mixed-quad output
Lets API users request triangle faces or a mixture of triangular and quadrilateral faces.
Three reduction levels
Provides high, medium, and low face-reduction settings for different quality and efficiency targets.
Asynchronous API jobs
Returns a job ID when work is submitted so an application can check the result separately.
Tencent Cloud SDK support
Works with Tencent Cloud's documented SDKs for Python, Java, PHP, Go, Node.js, .NET, and C++.
Broader Hunyuan 3D workflow
Sits alongside 3D generation, part generation, texture editing, UV unfolding, auto-rigging, and format conversion APIs.
Process
How the Hunyuan 3D Smart Topology (PolyGen) workflow works
Step 1
Prepare a high-poly asset
Supply a GLB or OBJ file by URL. Tencent recommends non-topologized high-poly inputs such as hard-surface models, characters, props, and everyday objects.
Step 2
Choose the mesh settings
Select triangle or mixed triangle-and-quad faces and, when needed, a high, medium, or low reduction level.
Step 3
Submit the API job
Call Submit3DSmartTopologyJob with Tencent Cloud credentials and retain the returned job ID.
Step 4
Retrieve and inspect the result
Poll the corresponding job-description endpoint, download the generated asset, and inspect silhouette, deformation areas, UVs, and edge flow before production use.
Cost
Hunyuan 3D Smart Topology (PolyGen) pricing and free plan
Smart Topology consumes 50 Tencent HY 3D Global credits per successful call. Credits cost $0.0135-$0.015 in prepaid packages or $0.02 postpaid, making one job roughly $0.68-$0.75 prepaid or $1 postpaid before optional concurrency fees.
First-time free credits
$0
New Tencent HY activations receive a one-time 200-credit resource package valid for one year.
- Equivalent to four 50-credit Smart Topology calls
- Does not automatically switch to postpaid unless enabled
- Availability is governed by the Tencent Cloud account and service terms
Prepaid credits
From $15
Buy a one-year package, starting with 1,000 credits for $15.
- 1,000 credits: $15
- 10,000 credits: $145
- 50,000 credits: $700
- 100,000 credits: $1,350
Postpaid
$0.02 per credit
Pay daily for usage after explicitly enabling postpaid billing.
- A 50-credit Smart Topology call costs $1
- Failed tasks are not charged
- Default Smart Topology concurrency is one job
Additional concurrency
$5,000/month
Add one concurrent processing slot for higher-volume pipelines.
- Priced per added concurrency slot
- Separate from per-credit usage charges
Pricing checked . Check current pricing at the source ↗
Assessment
Hunyuan 3D Smart Topology (PolyGen) strengths and limitations
Where it stands out
- Targets a specific production bottleneck instead of bundling retopology into a vague general-purpose generator.
- Supports both triangle and mixed triangle-and-quad topology.
- Has documented API inputs, job handling, SDKs, and transparent credit consumption.
- Fits alongside Tencent's generation and 3D post-processing services for automated pipelines.
What to consider
- The current service is designed around API integration, so it is less approachable than a visual desktop retopology workflow.
- It accepts only GLB or OBJ inputs by URL, with a documented 200 MB limit.
- Tencent warns that already-topologized or complex models may not support face reduction.
- Default concurrency is one job, while each additional concurrency slot is expensive.
- Automated topology still needs inspection around joints, silhouettes, UV seams, and other production-critical areas.
Compare
Hunyuan 3D Smart Topology (PolyGen) alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Design
Rodin
Consider Rodin when the larger need is generating detailed 3D assets from text or images rather than retopologizing an existing mesh.
Explore Rodin →Design
Tripo AI
Consider Tripo AI for a more creator-oriented hosted workflow spanning 3D generation and asset preparation.
Explore Tripo AI →Design
Meshcapade
Consider Meshcapade when the work centers on animatable digital humans and body models rather than general-purpose mesh topology.
Explore Meshcapade →Questions
Hunyuan 3D Smart Topology (PolyGen) FAQs
What is Hunyuan 3D PolyGen?
Hunyuan3D-PolyGen is Tencent's AI mesh-topology technology. Tencent Cloud's current Smart Topology API uses the Polygen 1.5 model to turn a high-poly 3D input into a neater, lower-polygon model.
Does PolyGen create a 3D model from text?
Not in the current Smart Topology workflow. It post-processes an existing GLB or OBJ model. Tencent offers separate Hunyuan 3D APIs for generating models from text, images, sketches, or multiple views.
How much does Hunyuan PolyGen cost?
A Smart Topology job uses 50 credits. That is roughly $0.68-$0.75 with current prepaid packages or $1 at the $0.02 postpaid credit rate. New activations receive 200 one-time free credits.
Does it support quad topology?
Yes. The API supports a triangle setting and a mixed setting containing triangular and quadrilateral faces.
Which files can I submit?
The official API documentation lists GLB and OBJ files supplied by URL, with a maximum file size of 200 MB.
Is the result automatically game-ready?
Treat it as a production accelerator, not an automatic approval step. A 3D artist should still inspect topology density, deformation zones, UV behavior, normals, materials, and compatibility with the target engine.
Bottom line
Our Hunyuan 3D Smart Topology (PolyGen) verdict
Hunyuan 3D Smart Topology is compelling when a team wants API-scale retopology for high-poly assets and can work inside Tencent Cloud. Its current Polygen 1.5 implementation is much more concrete than the original model announcement, but individual artists may find the API-first workflow and concurrency pricing less attractive than an interactive tool.
Visit Hunyuan 3D Smart Topology (PolyGen) website ↗Similar Design Tools
View all Design AI tools →
Qwen-VLo
Qwen-VLo - Alibaba’s 4o-style model for image generation and editing
Magic Animator
Magic Animator - Animate Figma designs in seconds with AI
Flux 1 Kontext
Flux 1 Kontext [dev] - BFL's Open-weight, SOTA image editing model
Gemini 2.5 Flash Image
Gemini 2.5 Flash Image - Google's SOTA AI image editing model
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.