Skip to content

Video generation

78 video models across text-to-video, image-to-video, video-to-video editing, and clip extension.

Quick start

bash
# text-to-video
gen-ai generate -m seedance-2.0 -p "a fox running through autumn leaves" -d 8

# image-to-video
gen-ai generate -m kling-motion-control-v3 -p "slow push in" -i ./portrait.jpg
json
{ "name": "picsart_generate",
  "arguments": { "model": "seedance-2.0", "prompt": "a fox running through autumn leaves", "duration": 8, "aspectRatio": "16:9" } }

Input types

TypeCommandModels
t2v text→videogen-ai generateSeedance 2.0, Kling V3, Sora 2 / Pro, Veo 3.1, Wan 2.7, Hailuo 2.3, LTX, Pika, Luma Ray 2
i2v image→videogen-ai generate -iKling Motion Control, Wan 2.7 I2V, Runway Gen4 Ref, Pika Scenes, VEED Fabric, HeyGen, Creatify
v2v video→videogen-ai generate --video / gen-ai extendSeedance Video Edit/Extend, Wan Video Edit, Runway Aleph, Grok Edit/Extend, LTX Extend/Retake, Sora 2 Extend

Providers

ProviderModelsHighlights
SeedanceSeedance 2.0 (+ Fast, Edit, Extend)Reference image, keyframes, native audio, 4–15s
GoogleVeo 3.1 / Fast / Lite1080p+, native audio, chainable extend
OpenAISora 2, Sora 2 Pro, Sora 2 ExtendPhotoreal motion + synced audio
KlingKling V3, Video O1, Motion Control, Avatar, EffectsHigh-motion; motion control from a still
WanWan 2.7 (t2v/i2v/r2v/edit)Versatile, crisp detail
RunwayGen 4.5, Aleph, Gen4 Ref, AvatarEditing & reference workflows
MiniMaxHailuo 2.3 (+ Pro/Fast)Fast, expressive motion
LTXLTX 2.3 Pro/Fast, A2V, Extend, RetakeAudio-to-video & re-takes
LumaRay 2, Flash 2, ReframeReframing & fast generation
PikaPika 2.2, Scenes, FramesScene & keyframe control
GrokImagine Video, Edit, Extend
VEED · HeyGen · CreatifyFabric, Talking Photo, AuroraAvatar / talking-photo
ByteDance · OVI · Videography · HappyHorseUpscaler, OmniHuman, OVI, Videography, Happy HorseSpecialized

Extending clips

bash
gen-ai extend --video ./clip.mp4 -p "the camera keeps panning right" --times 2

The Veo family extends +7s per segment; Seedance, Sora, Grok, and LTX expose dedicated extend models.

Common video parameters

ParamCLI flagNotes
prompt-pRequired (or stdin)
aspectRatio--are.g. 16:9, 9:16, 1:1
resolution-re.g. 720p, 1080p, 4k
duration-dSeconds (model-dependent set)
generateAudio--audio-genNative audio track
imageUrls / videoUrl-i / --videoi2v / v2v inputs

Built on @picsart/ai-sdk · gen-ai CLI · Picsart MCP · Skills