Text2Movie

The film language each AI video model's docs let you write

Which AI video models take an end frame?

14 of the 24 models: Adobe Firefly, Google Veo, Kling AI, LTX, Luma Dream Machine, Midjourney, MiniMax Hailuo, Moonvalley Marey, Pika, PixVerse, Runway, SkyReels, Vidu and Wan. The other 10 do not, in the docs read. As of 2026-09-25.

The register, split by the questionWhich AI video models take an end frame? 14 models match and 10 do not.RegisterMatch14 models: Adobe Firefly, Google Veo and Kling AI and moreNo match10 models
Fig. 1 Counted from the models' docs on this page.
Models that answer the question, read 2026-09-25.
ModelFramesWhat the docs say
Adobe FireflyFirst and last frameA first and a last keyframe image can be uploaded as fixed points for the clip
Google VeoFirst and last frameA separate mode generates video from a first and a last frame
Kling AIFirst and last frameStart and end frames to video is supported in VIDEO 2.6 and 3.0
LTXSeveral keyframesThe model supports several keyframes, not only a first frame
Luma Dream MachineFirst and last frameKeyframes set a start image, an end image or both
MidjourneyFirst and last frameThe image is the starting frame; --end sets an ending frame and --loop makes the two the same
MiniMax HailuoFirst and last frameA first-frame image, a last-frame image or both can be set
Moonvalley MareySeveral keyframesSeveral reference images are placed on a timeline and the model makes the transitions
PikaFirst and last framePikaframes takes a first and a last frame as still images
PixVerseFirst and last frameA Transition endpoint generates video between a first and a last frame
RunwaySeveral keyframesImage to video uses an input image; keyframes moved to an Animate Frames app
SkyReelsFirst and last frameStart and end frame control is part of the diffusion forcing model's inference code
ViduFirst and last frameTwo images set the start frame and the end frame
WanFirst and last frameA separate image-to-video mode takes a first and a last frame

Inclusion rule. Models whose own docs meet the condition in the question. Order. Alphabetical by model name.

1Why it matters

An end frame pins where a shot finishes, which is what lets it cut cleanly to the next shot. Models that take several keyframes go further and let the path between them be set too.

2From the docs

Along with the image for the first frame, you can also upload one for the Last frame to conclude the video.

Adobe, Firefly help: generate videos using images

First and last frames to video

Google Cloud, Veo video prompt guide

3The rest of the register

Not matching: Grok Imagine, Higgsfield, HunyuanVideo 1.5, Leonardo Motion 2.0, LTX Studio, Open-Sora 2.0, SceneMixer, Seedance, Sora and Step-Video-TI2V.

4Related questions