Create your own avatar

4 steps

1. 1. Choose a model

   You have	Use
   A photo of a real person	essence-2: photoreal, up to 1080p
   A character, animal, robot or illustration	expression-2: any subject
   Not sure	auto: people go to essence-2, everything else to expression-2

   Every model except Expression 2 needs a clear, real human face: see Choosing a model.

   Expected

   One model value: essence-2, expression-2 or auto.

   Done: step 1

2. 2. Pick the inputs

   Input	Use for	Limits
   Image	the face	under 10 MB; one clear figure, neutral expression, facing the camera, face unobstructed
   Voice	voice cloning	30 seconds or more of clean speech, one speaker, no music (MP3, WAV or M4A); the whole file is used
   Prompt	a description of the personality	required when there is no image

   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). For essence-2, the face also has to be big enough in the photo: a head-and-shoulders or chest-up photo works best, and a photo from the thighs up or wider may be turned away with 422 IMAGE_FACE_UNSUITABLE.

   The photo is checked before anything is charged; the refusals are in Troubleshooting. For an animal or character, use expression-2 with a well-lit, front-facing photo and the face filling the frame.

   Without a voice sample, a voice is generated to match the persona. Your prompt is a description: creation writes the avatar’s full system prompt from it, so the stored text is new and is not a copy of yours.

   Without a prompt, a role is picked at random from a built-in list and the persona is written from that and the image. To store exact text, create the agent, wait until it is ready, then update it with your system_prompt; an update is saved as sent.

   Expected

   An image URL that opens in a browser without signing in, under 10 MB, plus a voice sample and a prompt if you have them.

   Done: step 2

3. 3. Create the agent

   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": "YOUR_IMAGE_URL"}'

   The response carries an agent_id. Creation takes minutes for Essence 2 and longer for Expression 2. You can also create an agent in the Console at bithuman.ai, where you pick the model.

   Expected

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

   Done: step 3

4. 4. Wait until it is ready

   Poll GET /v1/agent/status/{agent_id} every few seconds until status is ready or failed.

   Expected

   status is ready: after a few minutes for Essence 2, or after a longer build for Expression 2.

   Done: step 4

All steps done.