Talking avatar API from one photo

Send one portrait to POST /v1/agent/generate and bitHuman creates a talking avatar from it in the bitHuman cloud. Then talk to the avatar in real time through the REST API, an SDK or the LiveKit plugin; the finished avatar runs anywhere its model runs.

Which model to use

Essence 2 renders a photoreal person from one portrait; Expression 2 renders any character (people, animals, cartoons) from one portrait.

Every model except Expression 2 needs a clear, real human face. Always send model: an omitted model creates a first-generation Expression 1 agent.

Create the avatar

curl -X POST https://api.bithuman.ai/v1/agent/generate \
  -H "Content-Type: application/json" -H "api-secret: $BITHUMAN_API_SECRET" \
  -d '{"model": "auto", "prompt": "You are a friendly receptionist.", "image": "https://example.com/headshot.jpg"}'

{"success": true, "message": "Agent generation started", "agent_id": "A80HVD8577", "status": "processing"}

Poll GET /v1/agent/status/{agent_id} every few seconds until status is ready or failed. An Essence 2 avatar is ready in minutes; an Expression 2 avatar takes longer to build. Every step, with the voice and the persona: Create your own avatar.

What makes a good photo

One subject, in focus, facing the camera with a resting expression and the whole face visible (eyes, nose and mouth). The image is under 10 MB, at a URL that opens in a browser without signing in. For an animal or character, use expression-2 with a well-lit, front-facing photo and the face filling the frame.

What it costs to create

Creating your own avatar is a one-time 500 credits for Essence 2 (ready in a few minutes) or 2,000 credits for Expression 2 (it takes longer to build).

A failed creation is refunded automatically. API use needs the Creator plan or higher. Every rate is on Pricing.

Talk to it in real time

To make a live avatar say a line from your backend, call POST /v1/agent/{agent_id}/speak.

A real-time session bills to the second at the rates on Pricing; rendering on your own hardware or on the device bills a lower rate than the bitHuman cloud.

Try a sample first

The wise-pup sample (Expression 2) runs in the web embed with no account or install:

<iframe src="https://www.bithuman.ai/embed/A23WJF0199" allow="microphone *" style="width:100%;height:600px;border:0"></iframe>

Open the page, allow the microphone and talk: the avatar greets you and answers. A sample session ends after a minute. The sample avatars are sofia-ramirez (Essence 2) and wise-pup (Expression 2): Sample avatars.

Troubleshooting