Workflows you can see, draw, and trust.
Ask in plain words. An agent writes the pipeline, draws it on the canvas, and runs it — video, dubbing, faceswap, business automation. Drag to change anything; watch every run live.
Three moves. That's the whole product.
Describe it
Type what you want — "dub this in Spanish", "swap his face and have him sniff the fish". The agent writes the pipeline and lays it out on the canvas.
Shape it
Drag steps around, swap models, test one node at a time. Every change is versioned — undo anything.
Let it run
On a schedule or on demand. Failures retry themselves; you see cost, duration, and output for every run.
Say what you want.
The agent writes the pipeline.
Change what he says — instead of “Why is the fish so expensive?” make it “Why does the fish smell so good?” Swap his face for Nadia from our character studio, then have him sniff the fish until his head explodes.
{
"name": "fish-smells-good",
"input": "clip_014.mp4",
"steps": [
{ "id": 1, "op": "transcribe",
"model": "whisperx",
"find": "Why is the fish so expensive" },
{ "id": 2, "op": "revoice", "model": "f5-tts",
"voice": "clone:speaker_1",
"text": "Why does the fish smell so good" },
{ "id": 3, "op": "lipsync",
"model": "latentsync", "audio": "$2.audio" },
{ "id": 4, "op": "faceswap",
"model": "facefusion",
"face": "studio://nadia" },
{ "id": 5, "op": "extend", "model": "wan-2.2",
"prompt": "sniffs the fish, head explodes" },
{ "id": 6, "op": "export",
"upscale": "real-esrgan", "fps": 60 }
]
}
The same power is one API call:
POST /v1/pipelines/design
turns a plain-language request into a valid pipeline your systems (or agents) can run —
see the pipelines guide.
One pipeline, drawn in plan view.
Media, workflows, agents.
Production-grade media pipelines plus general business automation, all as canvas nodes. Full model catalog →
Nothing runs in the dark.
Step-level traces, auto-retries, and a replay button when something needs you.