Image generate (text-to-image)

POST /api/image-generate20 quota units per call · cache hits free

Generate a PNG image from a text prompt using Black Forest Labs's Flux.1-schnell model via Cloudflare Workers AI. Fast 4-step generation, prompt-only API, no negative prompt or seed required.

Text-to-image generation via Flux.1-schnell — Black Forest Labs's distilled, 4-step variant of the Flux family. Returns a PNG image of the prompt. Useful for: - Quick visual scaffolding (mockups, illustrations, hero art). - LLM-side asset generation ("render a logo for…"). - Avatar / icon synthesis for accounts without a profile image. Flux.1-schnell is optimised for speed and trades some quality vs the full Flux.1-pro. For higher-quality output use the (separate, paid) Cloudflare Flux Pro endpoint via a custom wrapper. **Cost: paid Cloudflare Workers AI inference.** Image generation is the heaviest model in the catalogue — each call is several seconds and charges against your account's Workers AI budget plus 20 wasm-tool quota units per call.

Inputs

Name	Type	Default	Description
prompt*	string	—	Natural-language description of the image to generate.
steps	number (1…8)	4	Diffusion steps. Flux-schnell is distilled for 4 steps; higher values offer diminishing returns and longer latency.

Response

Modes: binary, json. Cache: yes (24h TTL).

Every response carries x-cache (HIT / MISS / BYPASS), x-cache-signature (stable across identical inputs), and the rate-limit headers listed below.

Quota & limits

Cost per call	20 units against the 10,000-unit monthly quota. Cache hits are free — only cache misses decrement.
Burst limit	200 requests per rolling minute, shared across all tools.
Max request size	`16 KiB` (base64 inflates file payloads ~33% — the limit applies to the decoded bytes).
Max response size	`8 MiB`
Timeout	120s wall clock.
Rate-limit headers	`X-RateLimit-Limit`, `X-RateLimit-Remaining`, `X-RateLimit-Burst-Limit` on every response; `X-Quota-Warning` once 80% of the monthly quota is spent; `Retry-After` on 429s.

Errors

Errors are JSON with an error string; schema failures add a details object (Zod's flattened field errors).

Status	When	Body
400	Input fails the tool's schema (wrong type, missing required field, malformed base64) or the file bytes can't be parsed as the expected format.	{ "error": "invalid inputs", "details": { "fieldErrors": { … } } }
401	Missing or invalid X-Api-Key / X-Account-Id headers (or no session when called from the browser).	{ "error": "unauthorized" }
404	Unknown tool id, or calling /api/* on the website hostname instead of api.xtract.bot.	{ "error": "not found" }
413	Request body over the tool's max request size, or an image header declaring more megapixels than the tool's pixel budget.	{ "error": "…exceeds maxRequestBytes…" }
415	Content-Type isn't application/json on the json-body transport.	{ "error": "unsupported content-type; expected application/json" }
429	Monthly quota or the 200/min burst limit exhausted. Check Retry-After and the X-RateLimit-* headers.	{ "error": "monthly quota exceeded", "monthRemaining": 0, … }
5xx	Conversion engine failure. Safe to retry; if it persists, the input is hitting a bug — please report it.	{ "error": "…" }

Code samples

Built from the logo example.


curl -X POST https://api.xtract.bot/api/image-generate \
  -H "Content-Type: application/json" \
  -H "Accept: application/octet-stream" \
  -H "X-Account-Id: $XTRACT_ACCOUNT_ID" \
  -H "X-Api-Key: $XTRACT_API_KEY" \
  -d '{
  "prompt": "A minimalist geometric logo for a tools website, blue and white, vector style",
  "steps": 4
}'