TurboClip API

Video production, reduced to an API call.

Most video APIs return a raw clip and leave the production to you. TurboClip returns the deliverable — narrated, captioned videos, UGC-style ads, slideshows, and images, ready to post. Typed schemas, a live test client, and the exact cost quoted before a credit moves.

Three requests. One finished video. Ads, slideshows, and images run the exact same shape — estimate, create, poll.

  1. 1

    Price it first — free. The response quotes the exact cost and names the models that will run.

    Request

    curl -X POST https://api.turboclip.ai/v1/video/estimate \
      -H "Content-Type: application/json" \
      -H "x-api-key: tc_sk_..." \
      -d '{
        "prompt": "The 12-3-30 treadmill workout, explained in 30 seconds",
        "settings": { "aspectRatio": "9:16", "videoDurationSeconds": 30 }
      }'

    Response · 200

    {
      "success": true,
      "data": {
        "sufficient": true,
        "estimate": { "total": 206 },
        "models": [
          { "role": "text", "name": "Opus 4.8" },
          { "role": "image", "name": "Nano Banana (3 Pro)" }
        ]
      }
    }
  2. 2

    Send it — same body on POST /v1/video. Credits reserve up front, and the quote IS the charge.

    Request

    curl -X POST https://api.turboclip.ai/v1/video \
      -H "Content-Type: application/json" \
      -H "x-api-key: tc_sk_..." \
      -d '{ ...same body... }'

    Response · 200

    {
      "success": true,
      "data": {
        "projectId": "3f2a1b4c-…",
        "jobId": "9b1c7e2d-…",
        "creditsReserved": 206,
        "creditsRemaining": 294
      }
    }
  3. 3

    Poll, then collect the finished MP4 — narrated, captioned, ready to post.

    Request

    curl https://api.turboclip.ai/v1/jobs/{jobId} \
      -H "x-api-key: tc_sk_..."
    
    curl https://api.turboclip.ai/v1/video/{projectId}/result \
      -H "x-api-key: tc_sk_..."

    Response · 200

    {
      "success": true,
      "data": {
        "status": "completed",
        "videoUrl": "https://…/final.mp4"
      }
    }
Predictable billing

Know the cost before you spend.

Step 1

Estimate

Every flow has a free /estimate twin. It quotes the exact cost and names the models that will run — before anything happens.

Step 2

Create & reserve

The create call reserves the quoted credits. The reservation IS the charge — no metering surprises, only failed jobs refund.

Step 3

Poll

GET /jobs/{id} every few seconds: queued → processing → completed. Failures release the reservation automatically.

Step 4

Fetch the file

Signed 24-hour links to the finished MP4, slides, or media file — download and post.

What you can ship

The entire studio, behind one key.

Videos

A topic or a script goes in; a narrated, captioned MP4 comes out. Image-based scenes or premium AI-video clips.

UGC ads

Plan the shot list, review every beat, render only what you approve. The presenter speaks natively to camera.

Slideshows

Carousel-ready slides from your copy — with one locked, consistent character across every image.

Ad-hoc media

One image or one raw clip, straight into the media library. No project, no timeline, no ceremony.

Your own files

A presigned upload handshake turns local files into reference images, start frames, and presenter shots.

Full catalog access

Models with machine-readable capabilities, styles, teams, balances — everything an integration needs to self-serve.

Questions

Good to know.

How do I get an API key?

Sign in and open your profile menu → API Keys. The full key (tc_sk_…) is shown once at creation — store it securely; only a masked version is retrievable afterwards.

What plan do I need?

API & MCP access are included on the Creator plan and up, checked on every request. One welcome mat: a never-subscribed account can plan UGC ads for free — the paid gate is the generate step.

How does billing work?

Generation reserves credits up front, and the reservation is the charge — only failed jobs refund. Every flow has a free /estimate twin that previews the exact cost and models first.

What are the rate limits?

30 requests per minute per API key. Beyond it you get a 429 with a Retry-After header.

Can I call the API from a browser?

Yes — api.turboclip.ai serves open CORS. That’s safe because auth is key-based, never cookie-based. It’s also what powers the Test Request button in the interactive reference.

Is there a machine-readable spec?

Yes: https://api.turboclip.ai/openapi.json — OpenAPI 3.1, generated from the same schemas that validate every request. Import it into Postman or generate a typed client.

What’s the difference between the API and the MCP?

Same engine, two doors. The API is for wiring TurboClip into your own code; the MCP lets an AI assistant (Claude, ChatGPT, Cursor) drive it in plain language — no code required.

Build with TurboClip

Your first finished video is one request away.

Open the reference, paste your key, and fire a real request from the page. What comes back is ready to post.

API & MCP access on Creator and up · spec at api.turboclip.ai/openapi.json