Latest media models and pricing are continuously updated Browse model catalog →

AGROUTER · Generative Media

A generative media platform built for products and Agents.

Use production-ready image and video models; deploy your own models or use dedicated compute when more control is needed.

Triptych of generated image, video, and creative editing results
ImageVideoEdit

One interface for the entire generative media workflow.

100+Media Models
3Usage
1Shared Quota & Records

01 · Choose Product

Start from existing assets or models.

Different control over models, code, and infrastructure suit different products.

MODEL APIs · Default Entry Point

Use media models directly.

Call optimized, auto-scaled, and production-ready image and video models. Send JSON, get media URL.

  • Get Started Fast
  • Pay by Output or Usage
  • No GPU Management Required
Browse Models →
SERVERLESS · Custom Deployment

Deploy Your Own Model.

Bring Your Model Weights, Container, or Existing Service. AGRouter Manages GPU, Queues, Scaling, and Observability.

  • Control Code and Weights
  • Auto Scale to Zero
  • Pay by Runtime
Learn More About Serverless →
COMPUTE · Dedicated Compute

Continuous Use of Dedicated GPU.

Ideal for Training, Fine-tuning, Batch Processing, and Long-running Workloads Requiring Stable Hardware Access.

  • Full SSH Control
  • Fixed Hourly Pricing
  • Single or Multi GPU
Learn More About Compute →

02 · Explore Model API

View One Category at a Time.

Select classification dimension first, then choose specific capsules; irrelevant options will be automatically hidden.

View all →
POST/v1/media/generate

Filter dimensions can be changed, but the API call remains consistent.

Copy request →

03 · How to Choose

Call model, deploy model, or manage dedicated compute?

ProductWhat You ControlScaling MethodBillingBest Suited For
ServerlessCode, Weights, ContainerAutomatic, from scratchBy RuntimePrivate Models, Custom Pipelines
ComputeFull GPU InstancesManual, Continuous RuntimeFixed Hourly PricingTraining, Fine-tuning, Batch Processing

Enter AGROUTER System

Infrastructure is not the end; deliverable work is.

Most customers start with Model APIs. Verified media capabilities can directly enter the product, or be equipped with language models, APIs, and Skills for Agents.

01 · Select SupplyModel API
Serverless
Compute
→
02 · Verify CapabilitiesQuality
Latency
Cost
→
03 · DeliveryProduct API
Media Workflow
Agent
Run Results and Cost Return to the Next Model Selection ↺

Three Steps to Start

From creating an account to generating the first media result.

Use the same account, same general quota, and same AGRouter API Key as language models.

01

Create Account

Register for AGRouter, your personal space is automatically created; invite team members when collaboration is needed.

YouTeam+
02

Top-up General Credit

Start from ¥50, the credit is shared among image, video, and language models.

¥50All ModelsBalance
03

Get API Key

Select a media model, send the first request, and obtain a deliverable media URL directly.

POST/v1/media/generate
model: "flux-studio"prompt: "glass botanical product"
200 OKimage_url →
01 · Explore Capabilities→Next Step

First, bring media capabilities into existing AI tools to complete your first real creation.

After verifying the model and workflow, then embed services, review rules, and delivery locations into the Agent.