Skip to content

feat(video): Add lucy-2-v2v model#98

Open
AdirAmsalem wants to merge 3 commits intomainfrom
feat/add-lucy-2-v2v-model
Open

feat(video): Add lucy-2-v2v model#98
AdirAmsalem wants to merge 3 commits intomainfrom
feat/add-lucy-2-v2v-model

Conversation

@AdirAmsalem
Copy link
Contributor

@AdirAmsalem AdirAmsalem commented Mar 4, 2026

Summary

Adds SDK support for the new lucy-2-v2v video-to-video editing model. Pairs with DecartAI/api#682.

Usage

Video editing with a prompt

const result = await decart.queue.submit({
  model: models.video("lucy-2-v2v"),
  prompt: "Transform the scene into a cyberpunk city",
  data: videoFile,
});

With a reference image

const result = await decart.queue.submit({
  model: models.video("lucy-2-v2v"),
  prompt: "Apply this style to the video",
  data: videoFile,
  reference_image: styleImage,
  enhance_prompt: true,
  seed: 42,
});

With only a reference image (no prompt)

const result = await decart.queue.submit({
  model: models.video("lucy-2-v2v"),
  data: videoFile,
  reference_image: styleImage,
});

At least one of prompt or reference_image is required. Both can be provided together.

REST API

POST /v1/jobs/lucy-2-v2v
Content-Type: multipart/form-data
Field Type Required Description
data file Video file to process
prompt string ✅* Text prompt for editing
reference_image file ✅* Reference image to guide the edit
seed number Seed for reproducibility
resolution string "720p" (default, only option)
enhance_prompt boolean Enhance prompt with AI (default: true)

*At least one of prompt or reference_image must be provided. Both can be used together.

Model specs

  • Resolution: 1280×720 (720p)
  • FPS: 20
  • Prompt + reference image: both allowed together (unlike lucy-restyle-v2v which is mutually exclusive)

Note

Low Risk
Additive SDK support for a new video model (registry + validation + tests) with no changes to existing model behavior; main risk is minor regressions in model schema/typing for queue submissions.

Overview
Adds SDK support for the new queue video model lucy-2-v2v, including model registry metadata (paths, 1280×720 @ 20fps) and a new Zod input schema.

The new schema allows prompt and/or reference_image together, enforcing that at least one is provided, and updates TypeScript model-specific input docs/types accordingly. Unit tests and AGENTS.md are extended to cover valid submissions and validation failures for lucy-2-v2v.

Written by Cursor Bugbot for commit c5f22a8. This will update automatically on new commits. Configure here.

@pkg-pr-new
Copy link

pkg-pr-new bot commented Mar 4, 2026

Open in StackBlitz

npm i https://pkg.pr.new/@decartai/sdk@98

commit: e3559e5

Copy link

@cursor cursor bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.

@AdirAmsalem AdirAmsalem changed the title Add lucy-2-v2v model to SDK feat(video): Add lucy-2-v2v model Mar 4, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant