Face ID API

Face ID vs character training

These solve the same problem at different price points. Face ID takes a single photo and works instantly, but the likeness is approximate — good enough for avatars, stylisation and one-off scenes. Character training takes 4–30 photos and about ten minutes, and produces a much tighter likeness you can reuse forever.

A useful rule: if the user is generating a handful of images right now, use Face ID. If they are building an ongoing library of one person, train a character.

Face ID

one photo, instant, approximate likeness

Character training

4–30 photos, ~10 minutes, tight likeness, reusable

Generate from a reference photo

The reference drives the face; the prompt drives everything else. Describe the person as well as the scene — hair colour, build, age — because the prompt controls those, not the photo.

curl https://api.fastgencloud.com/v1/images/face-id \
  -H "Authorization: Bearer $FASTGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
      "model": "face-id",
      "image": "https://example.com/reference.jpg",
      "prompt": "a woman with dark hair in a wool coat, walking through a rainy city street at dusk, cinematic lighting",
      "style": "realistic",
      "n": 2
    }'

Getting a good likeness

• Use a clear, frontal, well-lit reference with one unobstructed face
• Likeness is strongest on style "realistic"; anime and cartoon carry the person more loosely by design
• identity_strength (0.1–1.5, default around 0.9) trades likeness against expression — push it up for closer resemblance, down if faces look stiff
• Describe the person in the prompt too: the adapter supplies identity, not hair colour or clothing
• Generate n=2–4 and pick — face pipelines vary more between seeds than plain text-to-image

Parameters

model

face-id

image

reference photo — https URL or data URI

prompt

scene, pose, outfit, lighting — 1–2000 characters, verbatim

style

realistic (default), anime, cartoon

identity_strength

0.1–1.5, default ≈0.9

size

one of 1024x1024, 832x1216, 1216x832, 896x1152, 1152x896

n

1–4 images per request

seed

optional integer for reproducible output

How the API works

Every generation is a job. The POST returns 202 with a job id immediately, so your request thread is never blocked on a GPU. You then either poll GET /v1/jobs/{id} or register a webhook_url and receive a signed callback when the job finishes.

Send an Idempotency-Key header and a retried request replays the original job instead of charging you twice. Your wallet is debited when the job is submitted and refunded automatically if it fails.

curl https://api.fastgencloud.com/v1/jobs/$JOB_ID \
  -H "Authorization: Bearer $FASTGEN_API_KEY"

Pricing and billing

Pricing is prepaid and usage-based: you top the wallet up with a card and each job debits it. There is no subscription, no per-seat cost, and no monthly minimum.

Images above 1 megapixel are billed pro rata by pixel count — a 1.5 MP image costs 1.5x the listed rate. Failed jobs are refunded in full, automatically.