AI video and image model specs and features

This table lists what each AI video and image model in Twistavo AI Studio accepts and what one reference output costs: 31 video models and 33 image generation models, with clip lengths, resolutions, aspect ratios, the sound setting, required inputs, video extend, motion control, video edit and credits, then 3 video editing and motion models with their inputs and features.

Updated . Download: CSV · JSON. License CC BY 4.0, attribution "Twistavo".

Which AI video model supports start and end frames, reference images, extend and motion control

Video models first, then image models, in AI Studio order. Sort by any column.
Seedance 2.5video4304 to 30, every 1 s480p, 720p, 1080p16:9, 4:3, 1:1, 3:4, 9:16, 21:9sound setting: on or offyesyesyesup to 30 reference images, 30 images in total with framesup to 10 videosup to 10 audio tracks50nonono5 s clip3016.673.20
Seedance 2.5 Extendvideo4304 to 30, every 1 s480p, 720p, 1080pnot publishedsound setting: on or offyesnonoup to 30 reference images1 video, requiredup to 10 audio tracksnot publishedyes, forward or backwardnonon/anot publishednot publishednot published
Seedance 2.0video4154 to 15, every 1 s480p, 720p, 1080p, 4k16:9, 4:3, 1:1, 3:4, 9:16, 21:9sound setting: on or offyesyesyesup to 9 reference images, 9 images in total with framesup to 3 videosup to 3 audio tracks, with an image or a video12nonono5 s clip1934.282.05
Seedance 2.0 Fastvideo4154 to 15, every 1 s480p, 720p16:9, 4:3, 1:1, 3:4, 9:16, 21:9sound setting: on or offyesyesyesup to 9 reference images, 9 images in total with framesup to 3 videosup to 3 audio tracks, with an image or a video12nonono5 s clip1082.391.15
Kling 4.0video3303 to 30, every 1 s720p, 1080p16:9, 9:16, 1:1, 21:9not publishedyesyesyesup to 10 reference images, 10 images in total with frames1 video, 3 to 30 s eachnonot publishednonono5 s clip1082.391.15
Kling 4.0 Flashvideo3203 to 20, every 1 s720p16:9, 9:16, 1:1, 21:9not publishedyesyesnoup to 10 reference images, 10 images in total with frames1 video, 3 to 30 s eachnonot publishednonono5 s clip1292.861.37
Kling 3.0 Turbovideo3153 to 15, every 1 s720p, 1080p16:9, 9:16, 1:1not publishedyesyesnononononot publishednonono5 s clip651.440.69
Kling 3.0 Standardvideo3153 to 15, every 1 snot published16:9, 9:16, 1:1sound setting: on or offyesyesyesnonononot publishednonono5 s clip761.680.81
Kling 3.0 Provideo3153 to 15, every 1 snot published16:9, 9:16, 1:1sound setting: on or offyesyesyesnonononot publishednonono5 s clip861.910.91
Kling 3.0 4Kvideo3153 to 15, every 1 snot published16:9, 9:16, 1:1sound setting: on or offyesyesyesnonononot publishednonono5 s clip2585.722.74
Wan 3.0video4104 to 10, every 1 s720p, 1080p16:9, 9:16, 1:1not publishedyesyesnononononot publishednonono5 s clip651.440.69
Wan 3.0 Primevideo4104 to 10, every 1 s720p, 1080p16:9, 9:16, 1:1not publishedyesyesnononononot publishednonono5 s clip1182.621.25
Wan 2.7video4104 to 10, every 1 s720p, 1080p16:9, 9:16, 1:1not publishedyesyesnononononot publishednonono5 s clip651.440.69
Wan 2.6video5105 to 10, every 5 s720p, 1080p16:9, 9:16, 1:1not publishedyesnonononononot publishednonono5 s clip651.440.69
MiniMax H3video5155 to 15, every 1 s2K16:9, 9:16, 1:1not publishedyesyesnononononot publishednonono5 s clip861.910.91
MiniMax Hailuo 2.3video6106 to 10, every 4 snot publishednot publishednot publishedyesyesnononononot publishednononon/anot publishednot publishednot published
Grok Imagine Video 1.5video4104 to 10, every 1 snot publishednot publishednot publishedyesnono1 to 8 reference images, at least 1 requirednononot publishednonono5 s clip1934.282.05
FLUX 3 Videovideo5205 to 20, every 1 s720p, 1080p, 2k, 4k21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16sound setting: on or offyesyesyesup to 8 reference imagesup to 3 videosnonot publishednonono5 s clip2585.722.74
Gemini Omni Flashvideo4104 to 10, every 1 s720p16:9, 9:16not publishedyesnonoup to 7 reference images1 videononot publishednonono5 s clip1292.861.37
Gemini Omni Flash 1.1video3103 to 10, every 1 s360p, 720p, 1080p, 4k16:9, 9:16not publishedyesyesyesup to 8 reference imagesup to 3 videosnonot publishednonoyes, edit mode5 s clip1292.861.37
Gemini Omni Flash 1.2video3153 to 15, every 1 s360p, 720p, 1080p, 4k16:9, 9:16not publishedyesyesyesup to 8 reference imagesup to 3 videosnonot publishedyes, extend modenoyes, edit mode5 s clip972.151.03
Grok Videovideo1151 to 15, every 1 snot published16:9, 9:16, 1:1not publishedyesyesnononononot publishednonono5 s clip651.440.69
Grok Imagine Video 1.5 Litevideo1151 to 15, every 1 s480p, 720p, 1080p16:9, 9:16, 4:3, 3:4, 1:1, 3:2, 2:3not publishedyesyesnononononot publishednonono5 s clip430.950.46
Happy Horse Videovideo3153 to 15, every 1 s720p, 1080p16:9, 9:16, 1:1, 4:3, 3:4not publishedyesyesnononononot publishednonono5 s clip1082.391.15
Kling 3.0 Omni (Frames)video3153 to 15, every 1 snot published16:9, 9:16, 1:1sound setting: on or offyesyesyesnonononot publishednonono5 s clip541.200.57
Kling 3.0 Omnivideo3153 to 15, every 1 snot published16:9, 9:16, 1:1sound setting: on or offyesnonoup to 8 reference imagesnononot publishednonono5 s clip541.200.57
Kling 2.6 Videovideo5105 to 10, every 5 snot published16:9, 9:16, 1:1sound setting: on or offyesyesnononononot publishednonono5 s clip861.910.91
MiniMax H3 Maxvideo5155 to 15, every 1 s480p, 768p21:9, 16:9, 4:3, 1:1, 3:4, 9:16not publishedyesyesyesup to 8 reference imagesup to 3 videosup to 3 audio tracksnot publishednonono5 s clip1082.391.15
Seedance 2.0 Minivideo4154 to 15, every 1 s480p, 720p16:9, 9:16, 4:3, 3:4, 1:1, 21:9sound setting: on or offyesyesyesup to 8 reference imagesup to 3 videosup to 3 audio tracksnot publishednonono5 s clip430.950.46
Google Veo 3.1video484 to 8, every 2 snot published16:9, 9:16not publishedyesyesnononononot publishednononon/anot publishednot publishednot published
Google Veo 3.1 Litevideo484 to 8, every 2 snot published16:9, 9:16sound setting: on or offyesyesyesnonononot publishednononon/anot publishednot publishednot published
Soul 2imagen/an/an/a720p, 1080p9:16, 16:9, 4:3, 3:4, 1:1, 2:3, 3:2n/ayesnonononononot publishedn/an/an/a1 image20.040.02
Grok Imagine 2.0imagen/an/an/a2k1:1, 4:3, 3:4, 16:9, 9:16n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/an/anot publishednot publishednot published
Recraft 4.1imagen/an/an/a1k1:1, 4:3, 3:4, 16:9, 9:16n/ayesnonononononot publishedn/an/an/a1 image110.240.12
Z-Image Turboimagen/an/an/anot published1:1, 4:3, 3:4, 16:9, 9:16n/ayesnonononononot publishedn/an/an/a1 image20.040.02
3D Jutsu Anglesimagen/an/an/a1k, 1.5k, 2k1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9n/ayesnonoup to 10 reference imagesnononot publishedn/an/an/a1 image220.490.23
FLUX.2imagen/an/an/a1k, 2k1:1, 4:3, 3:4, 16:9, 9:16n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image90.200.10
FLUX 3 Imageimagen/an/an/a768sq, 1k, 1.5k, 2k, 4k21:9, 2:1, 16:9, 3:2, 7:5, 4:3, 5:4, 1:1, 4:5, 3:4, 5:7, 2:3, 9:16, 1:2n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image180.400.19
Flux Kontextimagen/an/an/anot published1:1, 4:3, 3:4, 16:9, 9:16n/ayesnonoup to 4 reference imagesnononot publishedn/an/an/a1 image130.290.14
GPT Image 2imagen/an/an/a1k, 2k, 4k1:1, 4:3, 3:4, 16:9, 21:9, 9:16, 3:2, 2:3, 4:5, 5:4n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image611.350.65
GPT Image 2.5imagen/an/an/a1k, 2k, 4k1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9, 27:16, 16:27, 9:8, 8:9, 4:5, 5:4n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image30.070.03
GPT Image 2.6imagen/an/an/a1k, 2k, 4k1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9, 27:16, 16:27, 9:8, 8:9, 4:5, 5:4n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image70.160.07
GPT Image 2.6 Flareimagen/an/an/a1k, 2k, 4k1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9, 27:16, 16:27, 9:8, 8:9, 4:5, 5:4n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image70.160.07
Grok Imageimagen/an/an/a1k, 2k1:1, 1:2, 2:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image90.200.10
Ideogram 4.5imagen/an/an/a1k, 2k1:1, 1:2, 2:1, 2:3, 3:2, 4:5, 5:4, 9:16, 16:9, 5:8, 8:5, 3:4, 4:3, 9:22, 22:9, 9:23, 23:9, 3:8, 8:3, 5:12, 12:5, 1:3, 3:1n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image541.200.57
Ideogram 4.6imagen/an/an/a2k, 4k1:1, 1:2, 2:1, 2:3, 3:2, 4:5, 5:4, 9:16, 16:9, 5:8, 8:5, 3:4, 4:3, 9:21, 21:9, 9:22, 22:9, 9:23, 23:9, 3:8, 8:3, 5:12, 12:5, 1:3, 3:1n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image180.400.19
Image Autoimagen/an/an/anot published1:1, 4:3, 3:4, 16:9, 9:16n/ayesnonoup to 14 reference imagesnononot publishedn/an/an/a1 image180.400.19
Kling Image 4.0imagen/an/an/a2k, 4k, 8k16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image180.400.19
Kling O1 Imageimagen/an/an/a1k, 2k1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9n/ayesnonoup to 10 reference imagesnononot publishedn/an/an/a1 image50.110.05
MiniMax H3 Imageimagen/an/an/a1k, 2k, 4knot publishedn/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image350.780.37
Nano Bananaimagen/an/an/anot published1:1, 3:2, 2:3, 4:3, 3:4, 4:5, 5:4, 9:16, 16:9, 21:9n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image90.200.10
Nano Banana 2.1imagen/an/an/a1k, 2k, 4k1:1, 3:2, 2:3, 4:3, 3:4, 4:5, 5:4, 9:16, 16:9, 21:9, 4:1, 1:4, 8:1, 1:8n/ayesnonoup to 8 reference imagesup to 3 videosnonot publishedn/an/an/a1 image130.290.14
Nano Banana 2 Liteimagen/an/an/a1k1:1, 3:2, 2:3, 4:3, 3:4, 4:5, 5:4, 9:16, 16:9, 21:9n/ayesnonoup to 14 reference imagesnononot publishedn/an/an/a1 image90.200.10
Nano Banana 2imagen/an/an/a1k, 2k, 4k1:1, 3:2, 2:3, 4:3, 3:4, 4:5, 5:4, 9:16, 16:9, 21:9n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image130.290.14
Nano Banana Proimagen/an/an/a1k, 2k, 4k1:1, 3:2, 2:3, 4:3, 3:4, 4:5, 5:4, 9:16, 16:9, 21:9n/ayesnonoup to 14 reference imagesnononot publishedn/an/an/a1 image180.400.19
OpenAI Hazelimagen/an/an/anot published1:1, 3:2, 2:3n/ayesnonoup to 16 reference imagesnononot publishedn/an/an/a1 image350.780.37
Seedream 5.0 Flashimagen/an/an/a1k, 1.5k, 2k1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9n/ayesnonoup to 10 reference imagesnononot publishedn/an/an/a1 image50.110.05
Seedream 5.5imagen/an/an/a1k, 1.5k, 2k1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9n/ayesnonoup to 10 reference imagesnononot publishedn/an/an/a1 image310.690.33
Seedream 4.5imagen/an/an/anot published1:1, 4:3, 16:9, 3:2, 21:9, 3:4, 9:16, 2:3n/ayesnonoup to 14 reference imagesnononot publishedn/an/an/a1 image90.200.10
Seedream 5.0 Liteimagen/an/an/anot published1:1, 4:3, 3:4, 16:9, 9:16, 21:9n/ayesnonoup to 8 reference imagesnononot publishedn/an/an/a1 image90.200.10
Seedream 5.0 Proimagen/an/an/a1k, 1.5k, 2k1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9n/ayesnonoup to 10 reference imagesnononot publishedn/an/an/a1 image220.490.23
Soul Castimagen/an/an/anot published16:9n/aoptionalnonononononot publishedn/an/an/a1 image20.040.02
Soul Cinematicimagen/an/an/anot published1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9n/ayesnono1 reference imagenononot publishedn/an/an/a1 image20.040.02
Soul Locationimagen/an/an/anot published1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9, 9:21n/ayesnonononononot publishedn/an/an/a1 image20.040.02

Editing, extend and motion control models

These video models have no generation duration and no reference cost in the catalogue, so only their inputs are listed.

Editing and motion models, in AI Studio order. Sort by any column.
Seedance 2.5 Edityesnonoup to 30 reference images1 video, requiredup to 10 audio tracksnot publishednonoyes, edits 1 uploaded video
Kling 3.0 Motion Controlyesyes, requirednono1 video, requirednonot publishednoyes, 1 character image and 1 motion videono
Kling 3.0 Motion Control Proyesyes, requirednono1 video, requirednonot publishednoyes, 1 character image and 1 motion videono

Feature definitions

Start and end frames
You upload the first image of the clip, and on some models the last one; the model animates from the first to the last.
Reference images
Images of a character, product or style that the model keeps in the clip without using them as the first frame. "Required" means the job does not run without them.
Video input
A video you upload for the model to follow, edit or extend. The cell gives the number of videos and, where the catalogue sets one, their length.
Audio input
An audio track you upload, which a speaking or singing clip follows. "No" means the model takes no audio track.
Sound setting
The AI Studio picker offers a sound setting for the model, on or off or always on. "Not published" means the catalogue has no sound setting for it.
Video extend
The model continues an uploaded video, adding seconds after it (forward) or before it (backward).
Motion control
The job requires a character image and a motion video; the model moves the character the way the person in the video moves.
Video edit
The model changes an uploaded video from a prompt, for example a background or an object, instead of generating a new clip.

How this table is produced

Durations, resolutions, aspect ratios, the sound setting, inputs and features are what the AI Studio picker offers for each model, read from the model catalogue: durations are the range the model's generation job accepts, narrowed by the per-job cost cap. The reference cost is the live AI Studio quote of a 5 s clip at the model's default resolution, sound on when the model has a sound setting and no reference media (the basis of the clip cost table), or of one image at default settings. Dollars convert credits at the $3.99 entry price. The best-plan column uses the Agency rate ($499/mo), the plan with the lowest price per credit. Inputs and features read the catalogue fields only (media roles and counts, required media, media limits, settings, the extension direction setting), never a model name; motion control is a job that requires both a character image and a motion video, and "no" means the field exists and offers nothing. Editing and motion models (video models with no generation duration and no reference cost) are listed apart, with their inputs only. "not published" marks a value the catalogue or the quote does not give; "n/a" a value that does not apply.

Cost of the same clip on every model · Every model's price · Open AI Studio