Build with the
Wireflow API
Design AI workflows visually, then call them from code. Pass a TikTok URL and get a full video breakdown. Send a prompt and get generated images. Any workflow becomes a programmable API endpoint.
# Run any workflow with just inputs
curl -X POST https://www.wireflow.ai/api/v1/workflows/YOUR_ID/run \
-H "Authorization: Bearer sk-your-api-key" \
-H "Content-Type: application/json" \
-d '{ "inputs": { "video_url": "https://tiktok.com/..." } }'
# Poll for results
curl https://www.wireflow.ai/api/v1/workflows/executions/EXEC_ID/poll \
-H "Authorization: Bearer sk-your-api-key"Three steps to automate anything
Build once in the visual editor. Call forever from code.
Discover inputs
GET /run returns the input schema — node IDs, labels, types, and defaults. Your code knows exactly what to pass.
GET /workflows/{id}/runRun the workflow
POST /run with just the input values. The server loads the saved workflow, injects your inputs, and executes the full pipeline.
POST /workflows/{id}/runPoll for results
GET /poll returns status and results. Poll every few seconds until COMPLETED. Long-running AI jobs finish reliably.
GET /executions/{id}/pollWhat people are building
Any visual workflow becomes a programmable API endpoint.
Video Intelligence
Pass a TikTok URL, get back scene-by-scene analysis with visual composition, content strategy, and recreation instructions.
Batch Image Generation
Send a list of prompts, get back generated images. Scale creative production without touching the UI.
Content Pipelines
Chain LLMs, image models, and video models into one workflow. Pass a topic, get back a full content package.
Brand Asset Creation
Upload a product photo, get back styled variations, social media crops, and video ads — all from one API call.
Video Remix Engine
Import any video, analyze every scene with AI vision, rewrite with a new theme, and regenerate frame by frame.
Multi-Model Orchestration
Route inputs through Claude for analysis, Flux for images, and Kling for video — all in one workflow execution.
Built for async AI
AI models take seconds to minutes. The API handles this gracefully with async execution and polling.
// Start the workflow
const { data } = await fetch(`/api/v1/workflows/${id}/run`, {
method: "POST",
body: JSON.stringify({ inputs })
}).then(r => r.json());
// Poll until complete
let delay = 1000;
while (true) {
const poll = await fetch(`/api/v1${data.poll}`)
.then(r => r.json());
if (poll.status === "COMPLETED") return poll;
if (poll.status === "FAILED") throw poll.error;
await sleep(delay);
delay = Math.min(delay * 1.5, 10000); // exponential backoff
}Everything you need
Production-ready infrastructure so you can focus on building, not plumbing.
Execute Any Workflow
Run image generation, video creation, audio synthesis, and multi-step AI pipelines with a single API call.
Async by Default
Submit jobs and poll for results. Long-running models like video generation complete reliably without timeout.
Scoped API Keys
Grant only the permissions each integration needs. Separate keys for reading, writing, and executing.
Plan-Based Rate Limits
Scale from 10 requests/min on free to 200/min on Enterprise. Standard headers tell you exactly where you stand.
Audit Logging
Every API call is logged with request IDs, latency, and credit usage. Debug and monitor your integrations.
Credit-Based Pricing
Pay only for what you use. Credits scale with your plan tier. Failed executions are automatically refunded.
API Endpoints
All endpoints are under /api/v1
/workflows/{id}/run/workflows/{id}/run/workflows/executions/{id}/poll/workflows/workflows/workflows/{id}/workflows/{id}/executionsStart building today
Create an API key and make your first request in minutes.