Provider Veed

VEED

Mode: video · Models: 2

Vendor: VEED · Official API docs: Fabric 1.0 API

VEED Fabric 1.0 is an image-to-video model that makes any still talk: an image plus an audio track produces a lip-synced MP4. There are no preset avatars — any photo, illustration, mascot, or 3D render works. Output is 480p or 720p, with duration following the audio length (up to ~5 minutes). It comes in a standard and a Fast variant.

Models

id	Name	Input type
veed-fabric-v1	VEED Fabric 1.0	i2v
veed-fabric-v1-fast	VEED Fabric 1.0 Fast	i2v

CLI

# make a portrait talk from a voiceover track
gen-ai generate -m veed-fabric-v1 \
  --image ./portrait.png \
  -a ./voiceover.mp3 \
  -r 720p

# faster, cheaper draft at 480p
gen-ai generate -m veed-fabric-v1-fast \
  --image ./mascot.png \
  -a ./speech.wav \
  -r 480p

MCP

{ "name": "picsart_generate",
  "arguments": {
    "model": "veed-fabric-v1",
    "imageUrls": ["https://example.com/portrait.png"],
    "audioUrl": "https://example.com/voiceover.mp3",
    "resolution": "720p"
  } }

Parameters — veed-fabric-v1

Param	CLI flag	Type	Values
imageUrls	-i (--image)	file	required — start image (1, JPG/PNG/WebP)
audioUrl	-a (--audio)	file	required — audio track; drives lip-sync and output duration
resolution	-r	enum	480p · 720p (default 720p)
prompt	-p	text	optional

Source: gen-ai models info veed-fabric-v1 --json. Duration is not a parameter — it equals the audio length (up to ~5 min). The Fast variant (veed-fabric-v1-fast) shares this surface.

Pricing

gen-ai pricing veed-fabric-v1 -r 720p

Cost is billed per second and scales with resolution (480p is cheaper than 720p) and the audio length, which sets the output duration.