santa hat
deerdeer nosedeer glow
Sign In
Video Generation November 2024 Update

UPDATE ! CivitAI added Mochi & Haiper Video on site Generation !
Don't forget all my LTXV, CogVideo & Mochi Video Prompt lists are here:
https://github.com/MushroomFleet/DJZ-Nodes/tree/main/prompts

November Update

Donut-Mochi V8 showcase video:

please comment if we missed your favorite AI video generator!


On Site

These models can now be run right here on CivitAI with BUZZ !

  • Mochi

  • Haiper

Workflows

These can be run "at home" with local GPU

These are often paid services with supercompute in the cloud

  1. Hailuo (minimax model) **NOVEMBER
    https://hailuoai.com/video

  2. Noisee.ai AMV generation
    https://www.noisee.ai

  3. txt2motion
    https://www.mootion.com/landing

  4. Pika.art video generation **NOV
    https://pika.art/

  5. Eleven Labs Dubbing
    https://elevenlabs.io/dubbing

  6. D-ID, Virtual Assistant Video
    https://studio.d-id.com/

  7. Haiper.ai, Video Generation
    https://app.haiper.ai/

  8. synclabs.so, video AI lipsync
    https://synclabs.so/

  9. lumalabs.ai, Video Generation JUNE
    https://lumalabs.ai/dream-machine/dream-machine

  10. Runway.ml, GEN-3 model **NOV
    https://app.runwayml.com/video-tools

  11. LTX.studio, Storyboard focused multi-generation service
    https://ltx.studio/ **JULY

LTXV

The fastest Local generation model released as of time of writing ( 23rd November 2024). Allows generation of photorealistic videos with Light Tricks LTXV model. My XODA-LTXV workflow pack offers text2video and image2video with prompts/images. I also provided a finetuned LLM PromptGen module for Claude & 110 prompts in 11 lists by category. It can deliver 2 seconds of video in ~8 seconds of generation time, this is the fastest Local GPU video generation model.

Mochi1

Allows generation of photorealistic videos with Mochi1. My Donut-Mochi workflow packs offered text2video and image2video with prompts/images, featured save/load latents and high quality mix of model/clip types to achieve the best outputs. It can run on a 12GB GPU which represents the most accessible video generation option for local GPU owners.

CogVideo

Allows generation of random animation with CogXVideo model. The workflow used text to video with a wrapper from Kijai.

SDV-XT 1.1

Allows generation of random animation with Stable Diffusion Video XT 1.1. The workflow used a single input image with no prompt.

SDXL Animatediff

Used existing SDXL Checkpoints (and lora) with Controlnet and Animatediff XL. The workflow requires a stack of models to run, and take a video input. full details are explained on the workflow page.


Local Above ~~~ Paid below

Noisee Animated Music Video Generation

Takes a song URL, Discord image URL and general style and/or scene by scene prompts. Produces 60 seconds of high quality 1024x576 video (aspect ratio is a parameter)

Mootion Text to Motion

Discord service allows us to select a 3D model from a list, while prompting animation. These animations can be transformed with prompts (and saved for use with animatediff)

Eleven Labs Dubbing Translator

Translate your content across 29 languages in seconds with voice translation, speaker detection, and audio dubbing, could be used as part of a multi stage project, but lip sync is a new service that pops up in an improved form from time to time.

D-ID Virtual Assistants / Agents

Generated Audio and video, lip synced with a personality, programmable with API

Haiper AI

Currently one of the best text2video HD & img2video HD generators out there, for music video creation

SyncLabs

Allows you to redub a video with AI lipsync

LumaLabs

Best known for their motion capture AI, they recently released a groundbreaking text and image to video service.

RunwayML

the original trainers of SD1.5, they have released Gen-3 Video Generation service as of July 2024.

also allows you to redub videos with AI lipsync

LTX Studio

Storyboard based AI video generation, used LLM to expand your prompt and allows later prompting in each scene, many options for soundtrack and speech generation, can output as a finished story.

other methods will be added here soon :) please comment with services we missed.


86

Comments