What is TryOn-API.com?

TryOn-API.com is the OpenRouter for virtual try-on: a single, OpenAI-compatible API gateway that routes virtual try-on and image-generation requests to many underlying models (Kling, Google Gemini, OpenAI gpt-image-1, FAL FLUX) behind one key, one schema and one unified credit balance.

Is the API OpenAI-compatible?

Yes. The /api/v1/chat/completions route follows the OpenAI and OpenRouter schema, including streaming via SSE and modalities:["image"]. Existing OpenAI SDKs work by changing the base URL to https://tryon-api.com/api/v1.

How do I migrate from OpenRouter or OpenAI?

Change three things: set the base URL to https://tryon-api.com/api/v1, use your TryOn-API key, and pick a try-on model slug such as kling/kolors-v1-5. No new SDK is required.

Which try-on models are supported?

kling/kolors-v1-5 (dedicated apparel try-on), google/gemini-2.5-flash-image, openai/gpt-image-1, fal/flux-kontext, and wearfits/tryon-clothing (dedicated apparel try-on with reusable digital twins). The live list is available at GET /api/v1/models.

How is pricing calculated?

Per successful request, in credits, drawn from one unified balance that works across every provider. Most models cost 1 credit per request; gpt-image-1 costs 2. Check your balance at GET /api/v1/credits.

How fast is a try-on?

Generative-image models (Gemini, gpt-image-1, FLUX Kontext) typically return in about 5 to 25 seconds synchronously. The dedicated Kling Kolors try-on runs about 30 to 60 seconds via submit-and-poll.

Can I run a try-on without the chat API?

Yes. POST /api/v1/tryon accepts a simple multipart request with person and garment files. Synchronous models return the image directly; asynchronous models return a job id you poll at GET /api/v1/tryon/{id}.

{Unified try-on router}

One API for every virtual try-on model

TryOn-API is the OpenRouter for virtual try-on. Route to Kling, Gemini, gpt-image-1, FLUX Kontext and more through a single OpenAI-compatible endpoint — unified billing, one key, zero lock-in.

Open Playground Read the docs Browse models

14models

7providers

1endpoint

OpenAIcompatible

            
              
            
            tryon.ts
          
# Any OpenAI-compatible SDK — change three things:
const client = new OpenAI({
  baseURL: "https://tryon-api.com/api/v1",   // 1. base URL
  apiKey:  process.env.TRYON_API_KEY,        // 2. your key
});

const res = await client.chat.completions.create({
  model: "kling/kolors-v1-5",                // 3. a try-on model
  modalities: ["image"],
  messages: [{ role: "user", content: [
    { type: "text", text: "put the garment on the person" },
    { type: "image_url", image_url: { url: personUrl }, tryon_role: "person" },
    { type: "image_url", image_url: { url: garmentUrl }, tryon_role: "garment" },
  ]}],
});
const url = res.choices[0].message.images[0].image_url.url;

Why TryOn-API

One endpoint, many models

Dedicated try-on (Kling Kolors) and generative image models (Gemini, gpt-image-1, FLUX) behind one schema.

OpenAI-compatible

Chat completions, streaming SSE, modalities:["image"]. Drops into your existing SDK.

Unified credits

One balance across every provider. Transparent per-request pricing in credits.

Native or chat

Use the OpenAI-style chat route, or a simple multipart /tryon with person + garment files.

Console

One API for every virtual try-on model

One endpoint, many models

OpenAI-compatible

Unified credits

Native or chat

Models

Rankings

Playground

API Keys

Credits

Docs