Repolode
Submit
tin2tin

tin2tin/Pallaidium

A free, open-source generative AI movie studio that runs inside Blender's Video Sequence Editor, covering text, image, video, audio and speech generation.

Open-source alternative to:Runway

View on GitHubWebsite
Stars
1,550
Forks
159
Open issues
4
Last commit
5 days ago

What is it?

Pallaidium is a Blender add-on that adds generative AI tools to the Video Sequence Editor. Users can turn prompts or existing strips into images, video, music, speech and text, with output placed back on the VSE timeline as new strips. The add-on is aimed at prototyping a film with AI from script or lyrics to generated shots, voices and music before shooting real footage.

The add-on supports a broad set of models, including FLUX, Qwen-Image, LTX-Video, Wan, Chatterbox and MOSS-TTS, and it discovers models from a plugin folder. It also offers features such as ControlNet, LoRA loading, IP-Adapter, a render queue with job status tracking, and batch processing of strips. Optional remote backends let generation run on ComfyUI, fal.ai or any server that follows an OpenAI-style /v1 contract.

Pallaidium targets Windows with an Nvidia CUDA GPU; Linux support is limited, and the models require substantial VRAM and disk space, with each model often several gigabytes.

Key features

  • 01Generates images, video, music, speech and text from prompts or Blender strips.
  • 02Adds results back to the VSE timeline as new strips with seed and prompt in the name.
  • 03Supports ControlNet, OpenPose, Canny, IP-Adapter and multiple LoRAs.
  • 04Includes a render queue with PENDING, RUNNING and COMPLETED job status.
  • 05Can use ComfyUI, fal.ai or OpenAI-style servers as optional remote backends.

Use cases

  • Prototyping a short film or storyboard with generated shots before real production.
  • Generating voiceovers, dialogue or music for a Blender video edit.
  • Creating subtitles from speech with word-level timing into VSE text strips.
  • Batch-generating or upscaling many clips from a set of strips.

Topics

#generative-ai#blender#video-generation#text-to-image#text-to-speech#diffusion#video-editing#personal-agent#ai#aicinema

FAQ

What is Pallaidium?
A free, open-source generative AI movie studio that runs inside Blender's Video Sequence Editor, covering text, image, video, audio and speech generation. Pallaidium is a Blender add-on that adds generative AI tools to the Video Sequence Editor. Users can turn prompts or existing strips into images, video, music, speech and text, with output placed back on the VSE timeline as new strips. The add-on is aimed at prototyping a film with AI from script or lyrics to generated shots, voices and music before shooting real footage.
Is Pallaidium free to use? What license does it have?
Pallaidium is open source under the GPL-3.0 license. Check the license terms before using it in commercial products.
Is Pallaidium still maintained?
Actively maintained. The last commit to Pallaidium was on Oct 17, 2026.
What can I use Pallaidium for?
Prototyping a short film or storyboard with generated shots before real production. Generating voiceovers, dialogue or music for a Blender video edit. Creating subtitles from speech with word-level timing into VSE text strips. Batch-generating or upscaling many clips from a set of strips.
Where can I get Pallaidium?
The source code and installation instructions are on GitHub: https://github.com/tin2tin/Pallaidium

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects