Models

GPT 2

Stable

Strong for accurate prompt details, product-style shots, clean edits, and multi-reference composition (up to 8 reference images). Best when prompt fidelity matters most.

Product id

gpt-2

Price

$0.08 / call

Output

image (jpeg)

Avg latency

~60s

Parameters

Inputs

Send these in the JSON body of a POST /v1/generations request alongside "product": "gpt-2".

promptstringRequired

Text prompt that describes what to generate. 1–4000 characters. Be specific — subject, lighting, lens, mood, action.

negative_promptstringOptional

Things to avoid (e.g. `blurry, watermark, text`). Keep it short and concrete.

aspectstringOptionalDefaults to portrait

Output aspect ratio. `portrait` is best for character shots; `landscape` for scene compositions.

seedintegerOptional0 – 4,294,967,295

Deterministic seed. Same seed + same prompt = same image. Omit for a random seed.

webhook_urlurlOptional

HTTPS callback fired when the job reaches a terminal state. HMAC-SHA256 signed. See the Webhooks guide.

metadataobjectOptional

Arbitrary JSON object echoed back on retrieval and in the webhook. Useful for correlating jobs to your records. ≤ 4 KB.

POST /v1/generations — full reference for the create endpoint.
Playground — try this model inline with your key.

Quick call

POSTGPT 2/v1/generations

curl https://api.kiara-api.com/v1/generations \
  -H "Authorization: Bearer $KIARA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "product": "gpt-2",
       "prompt": "your prompt here"
     }'