OVI

Mode: video · Models: 1

Vendor: OVI · Official API docs: Fal — fal-ai/ovi

OVI jointly generates video and synchronized audio in a single pass — full scene plus matching audio, including spoken dialogue. It runs text-to-video by default and switches to image-to-video when a start image is supplied. Clips are short (~5s) with no length controls; dialogue and sound effects are described inline in the prompt via <S>…<E> (speech) and <AUDCAP>…<ENDAUDCAP> (audio caption) markup.

Models

id	Name	Input type
ovi	OVI	t2v

CLI

# text-to-video with a spoken line and SFX (joint audio)
gen-ai generate -m ovi \
  -p "A knight raises a sword. <S>For honor!<E> <AUDCAP>steel clang, distant crowd roar<ENDAUDCAP>" \
  --size 16:9

# image-to-video: animate a start image (shape follows the image)
gen-ai generate -m ovi -p "the portrait blinks and smiles" -i ./portrait.jpg

MCP

{ "name": "picsart_generate",
  "arguments": {
    "model": "ovi",
    "prompt": "A knight raises a sword. <S>For honor!<E> <AUDCAP>steel clang, distant crowd roar<ENDAUDCAP>",
    "size": "16:9"
  } }

{ "name": "picsart_generate",
  "arguments": {
    "model": "ovi",
    "prompt": "the portrait blinks and smiles",
    "imageUrls": ["https://example.com/portrait.jpg"]
  } }

Parameters

Full parameter surface for every model, sourced from gen-ai models info <id> --json. CLI flags show the primary short form; the canonical --kebab-case long form always works too.

ovi — OVI

Try ovi in Playground ↗

Input type: t2v

Param	CLI flag	Type	Values
prompt	-p	text	required
size	--size	enum	9:16 · 16:9 · 1:1 · 9:16+ · 16:9+ · 2:5 · 5:2 (default 16:9)
imageUrls	-i	file	image (up to 1)

Notes: size maps to OVI’s fixed WIDTHxHEIGHT resolution enum; it applies to text-to-video (image-to-video derives shape from the start image).

Pricing

gen-ai pricing ovi

Clip length is fixed (~5s) with no duration or resolution-tier controls, so OVI is a flat per-generation cost.