FigureLabs
Developer API

Scientific figures, built for your product.

Generate publication-ready figures from text, sketches, or reference images—all through one API.

View API docs
create-figure.sh
curl https://api.figurelabs.ai/v1/images/generations \
  -H "Authorization: Bearer $FIGURELABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "generations",
    "prompt": "PI3K–AKT signaling pathway"
  }'
202 Accepted · tsk_01JZ9F7KQ2 · processing

One API. Multiple workflows.

InputPrompt
Text-to-Figure input example
OutputScientific figure
Text-to-Figure output example
POST /v1/images/generations
{
  "mode": "generations",
  "prompt": "PI3K–AKT signaling pathway"
}
202 Accepted
{
  "task_id": "tsk_01JZ9F7KQ2",
  "status": "processing"
}
One endpoint. Automatic routing.View examples

Create, refine, and deliver scientific figures.

Describe your goal and provide any source files. FigureLabs automatically routes each request to the right generation or editing workflow.

Text-to-Figure

Create a figure from a scientific prompt.

Sketch-to-Figure

Turn a rough sketch into a polished visual.

Reference Image-to-Figure

Create from a reference image or visual style.

Enhance

Improve clarity, labels, and composition.

Flowchart

Generate workflows and process diagrams.

Upscale

Export at 2K, 4K, or 8K resolution.

Vectorize

Convert raster figures to editable SVG.

Quickstart

Submit, poll, and retrieve

FigureLabs uses asynchronous tasks so your product stays responsive while complex figures are generated.

  1. 1Submit a generation request.
  2. 2Persist the returned task_id.
  3. 3Poll until the task reaches a final state.
  4. 4Download the output URL within seven days.
import os, time, requests

headers = {
  "Authorization": f"Bearer {os.environ['FIGURELABS_API_KEY']}"
}
task_response = requests.post(
  "https://api.figurelabs.ai/v1/images/generations",
  headers=headers,
  json={"mode":"generations","prompt":"PI3K–AKT pathway"}
)
task_response.raise_for_status()
task = task_response.json()

while True:
  result_response = requests.get(
    f"https://api.figurelabs.ai/v1/tasks/{task['task_id']}",
    headers=headers
  )
  result_response.raise_for_status()
  result = result_response.json()
  if result["status"] in {"succeeded", "failed", "rejected", "canceled"}:
    break
  time.sleep(3)

if result["status"] != "succeeded":
  raise RuntimeError(result.get("error") or result["status"])
print(result["output_url"])