Plugins

Disclaimer: Steam-version does not have built in plugins, you need to install those manually: https://github.com/Lyric-Video-Studio/BuildInPlugins/

This is curated list of available plugins. All cloud based plugins require you to have your own access token. The app itself does not provide credits / access token to any 3rd party services. A1111-plugin does not required access tokens, but it does however require Nvidia RTX30xx series GPU and some technical know-how.

General advice that applies to all cloud based plugins: If you wish to regenerate item, clear out the “polling id”, otherwise it will just fetch the same image/video again.

Plugin feature / price sheet / overview

Company	Model	Fps	Max resolution	Max length (s)	Cost 720p/5sec
Luma AI	Ray 1.6	24	720p	9	0.35$
	Ray2	24	4K	9	0.7$
	Ray2-Flash	24	4K	9	0.24$
Runway ML	gen3a_turbo	24	720p	10	0.25$
	gen4_turbo	24	720p	10	0.5$
Minimax	S2V-01	25	720p	5	0.65$
	T2V-01	25	720p	5	0.43$
	T2V-01-Director	25	720p	5	0.43$
	I2V-01	25	720p	5	0.43$
	I2V-01-Director	25	720p	5	0.43$
	I2V-01-live	25		5	0.43$
Fal ai	see	the	link	below

Updated 18/04/2025, always check sources for changes… Sources: https://lumalabs.ai/api/pricing, https://docs.dev.runwayml.com/usage/billing/, https://www.minimax.io/platform/document/Pricing%20Overview?key=67373ec8451eeff1a85b9e4c, https://app.klingai.com/global/dev/document-api/productBilling/prePaidResourcePackage For fal ai plugin, see the price of models here: https://fal.ai/explore

Plugin interface

https://github.com/Lyric-Video-Studio/PluginInterface (for developers)

Runway ML

Runway ML offers img2vid services, text/image to iage and video-to-video. It works similarly than Luma Labs Dream machine, but you need to input starting frame. Ending frame not supported. This plugin supports uploading the source file to your DropBoxaccount, so you need not to worry about extra hazzle. Api keys can be obtained here: https://dev.runwayml.com/

ACT2 is now supported. For that, you need reference image of how the video should look like and reference video of what movements the the video should follow. Reference video can’t be bigger than 32mb (use trimmig to scale down or shorten the video if necessary)

For creating images, you need to have at least one reference image as well as tav for the character. The way it works is that you reference the characted in the prompt. See the example picture below. Tag there is ‘orc’ and in prompt it is referenced as ‘@orc’

Luma Labs Dream Machine

Create beautiful videos with Luma Labs Dream Machine. You can extend videos by copying pollingId from previous video. Also start / end frame is supported, along with txt2Vid. Support content delivery via your DropBox account. Now also supports RAY-2 video model. Key your API key from here: https://lumalabs.ai/dream-machine/api/keys

To upscale you generation, create video plugin track “Luma AI DreamMachine generation upscale”. Right click on any video generated with DreamMachine, select “Copy content id”. Create new item to upscale track. Paste the id to “Generation id” field, select target resolution and click “Generate”. Note, this workflow will be patched to be more user friendly / support importing directly from target track.
Note: Luma AI does not seem to work with image-to-video if Google Drive is selected as content delivery method.

Modify video is now supported: use reference image and reference video to create re-acted version of you video in your style of choice. Reference video is currently limited to 10sec (use trimmig to scale down or shorten the video if necessary)

Integrated Stable Diffusion

See https://lyricvideostudio.wordpress.com/help/#sd

FramePack

FramePack integration is available for local video generation. FramePack offers local video generation with 6gb of GPU ram. Although it’s very slow with that. You can generate practically limitless length videos.

First, install the FramePack locally, either from this link or click the link in ui. After that, setup the plugin to point to run.bat if your installation and you should be ready to go. LVS will start the FramePack when needed. Then, just prompt. During generation, FramePack makes 1 sec videos, that are displayed in timeline and you can cancel the video if it does not look right. Only image-to-video supported.

If you wish to use Framepack F1 (forward-version), just modify the run.bat from Framepack installation, change
“%DIR%\python\python.exe” demo_gradio.py –server 127.0.0.1 –inbrowser
TO
“%DIR%\python\python.exe” demo_gradio_f1.py –server 127.0.0.1 –inbrowser

WAN 2.2 Local / WAN api

WAN local mode is supported in full / trial versions, WAN API also in LITE version. In local mode, you first have to click select the folder where to store models and also click ‘Install requirement’. While installing, you should also click on ‘Download models’. Note that models take a huge amount of space, like 100’s of gigabytes and all the different models need to be downloaded separately (like text-to-video vs imahe-to-video. Minimum on 16gb GPU memory needed, much more recommended 🙂 Currently generation speed is quite slow, tens of minutes for 5sec clips, since it’s pretty new model. Updates to local mode will hopefully be available.

For API mode, which also takes several minutes per generation, you can use Alibaba cloud. It’s bit more trickier (and sketchy) to set up, compared to others, because they require credit card to be added, without any pre-paid amount of credits. You can set up spending limit alarm, but I haven’t yet found out how to set up hard limits. So, beware… You need to create your API key in Alibabab Model studio: https://modelstudio.console.alibabacloud.com/?tab=globalset#/efm/api_key

Since WAN 2.2 uses lot’s of memory and generation times are unpredictably long, you need to manually release the used memory after you are no longer generating more videos. Using for example stable diffusion or framepack at the same time WILL NOT work!!! Using API based services however works, while waiting for results:

Black Forest Labs

Create images, imgToImg and now also edit images with Black Forest Labs FLUX. Just register, add credits and grap your API key from https://dashboard.bfl.ai/keys Check out pricing from here: https://bfl.ai/pricing/api

Minimax / Hailuo

Get your api key & add credits in here: https://www.minimax.io/platform/user-center/basic-information/interface-key
MiniMax plugin supports text-to-video, image-to-video, subject-to-video, text-to-speech and music. Make sure you select the correct model when generating items. When using text, choose model starting with T2V. For images, I2V. You can add subject reference also. Then you must use S2V model.

For music creation, up to 600 character of lyrics is supported and separate prompt guides the music creation. To lyrics, you can add these tags to represent the parts of the song: [intro][verse][chorus][bridge][outro] More info

Minimax plugin also supports image generation, with character reference:

Chatterbox / Resemble.ai

Chatterbox can be run locally for text-to-speech purposes (non-LITE versions). Even with CPU, you can generate speech from text. Plugin also supports Resemble.ai API for text-to-speech (also in LITE version). When run locally, external software packages will be downloaded and installed to create sandboxed python environment for running the Chatterbox. Remember to also download models, after you have defined to folder where to store them. You can get the access tokes from here: https://app.resemble.ai/account/api

Google

Google plugin support gemini-2.5-flash-image-preview model, AKA nano banana. You can create and edit images with it. Get your api key from https://aistudio.google.com/apikey You can even generate images with free tier, with some rate-limitations. See https://aistudio.google.com for more details

MusicGPT

App now support generating music using MusicGPT api. Access tokens can be found here: https://musicgpt.com/api-dashboard Notice that their pricing varies based on the monthly subscription tier. They also offer “pay-as-you-go”-pricing, but it’s quite expensive. So, monthly subscription is recommended if you use it a lot. Note that each generation creates two audio outputs. You can switch between versions with “Switch to alternative”-button

ElevenLabs

ElevenLabs plugin supports turning text to audio narrative. You can choose from pre-made voices. Refreshing the voices happen automatically on first boot, but you might want to peridiocally refresh then by opening ‘General settings’ from plugin edit menu and clicking “Refresh voices”. ElevenLabs also seems to be providing generous free tier, 10k credits/month https://elevenlabs.io/pricing/api

Mistral AI

Mistal AI plugin can create images from text. It differs from other plugins so that you must specify that you want to create image in the prompt. Upon first image creation, AI agent is created for the image tasks. If ou have custom made ai agent in mistal AI, you can paste the agent id to General Settings. Note that image quality is not that high compared to others, but it’s fast and allows generation for free:

FalAi

With Fal.AI, you can use multiple models with single access token. NOTE: currently it looks like DropBox does not work as content delivery method in their service. Get your API keys from https://fal.ai/dashboard/keys NOTE x2: Not all models have been integrated from Fal.Ai to Lyric Video Studio. Only the models currently not available from official sources are integrated. More models will pop up when such are released. Currently supported models: sora-2, veo3, hailuo-02, wan-25 , v2.2-a14b, kling-video v2.5-turbo, ltxv-13b-098-distilled, pixverse, lucy-edit, omnihuman v1.5, seedvr, qwen-image, imagen4,hidream-i1-full, seedream

A1111

Locally hosted A1111 instance for machines that have high-end GPU. Manully install A1111 (https://github.com/AUTOMATIC1111/stable-diffusion-webui) and models and set the path to webui-user.bat in installation folder from settings. Plugin will insert needed command line arguments when starting the A1111. If you want to launch A111 manually, you must use –api –nowebui command line arguments, otherwise the API does not work

OpenAi / Stability.Ai

These plugins use cloud services to generate images / video. For OpenAi, get your API key from https://platform.openai.com/api-keys

For stability.ai: https://platform.stability.ai/account/keys

Do not share your access token publicly, because with the token, anyone can use your account to generate images/videos. Token are stored in secure store (https://learn.microsoft.com/en-us/dotnet/maui/platform-integration/storage/secure-storage?view=net-maui-8.0&tabs=windows) and not with your project file or plugin default settings. Each plugin is however responsible to keep access tokens secured.

Image crop / resize

This plugin does not use generative AI, but rather does scaling and cropping the “old school”. This plugin is useful when you to Img2img or ImgToVid, because ImgToVid is usually especially picky about the image size it accepts. With this plugin, you can either crop or downscale images if needed. Usual workflow goes like this:
1. Create images with any plugin
2. Create Image cropper plugin track and import images from step 1
3. Crop to desired size
4. Create Stability AI ImgToVid track
5. Import images from cropper track
6. Generate video

3rd Party plugins

Have you made a plugin and wish to feature it here? Contact me! Also, if you’re planning to create a plugin and would like to get the full version of the app, use contact form and describe your plans and I’ll add you to “beta-build channel” / flight delivery in Microsoft Store

Example plugins: https://github.com/Lyric-Video-Studio/BuildInPlugins
Interfaces: https://github.com/Lyric-Video-Studio/PluginInterface