Yue2 allows for custom voice training on a production-class music model for the first time. Sample by gueykhalamari trained w/ AI-Toolkit using Samuel L. Jackson’s voice, link below.

You can find the workflow by AbleJones in the minimax_h3_resources channel here: discord.gg/dstjQYQNt

Latent mask inpainting with H3 enables highly divergent changes while holding consistency. Examples by PhoenixRisen + AbleJones - workflow below.

You can find a similar workflow here: github.com/seitanism/ComfyUI…

Seamless video/audio continuations in MiniMax H3 via latent masking, demo'd by AbleJones:

Link: huggingface.co/Cseti/LTX2.3-…

CrossViewWarp allows you to change the camera arc of an existing video while maintaining the motion. Demonstration w/ LTX2.5 by @ChetiArt, link below.

seitanism injecting frames into specific positions with h3 - see how it handles 1 frame every 1 second:

(gen by orabazes)

Wan Animate 2 allows for highly versatile motion transfer in a wide variety of styles. Examples by Kijai:

On out-of-distribution requests, H3 can produce highly coherent visuals and audio - example by @nebsh83:

How your 6 y/o potato GPU reacts when you try to run Minimax (gen'd by _kruss_)

Inpainting based on a reference w/ Minimax H3. Examples by AbleJones, feat. @PurzBeats:

76 five-second clips exploring different animation styles with MiniMax H3. All generated locally on a six-year-old GPU by the_shadow_nyc:

The next generation of vid2vid will go to some magnificent, disturbing places (generated locally by Dawnll using Minimax M3)

Historically, the lag between a frontier video model launching and a broadly comparable open-weight model running locally has averaged roughly nine months. If that trend holds, we should expect a Seedance 2-level local model sometime between August and December this year.

I have only ever banned spammers and people who were being extremely negatively, I harbor no ill-will against you but you can use the bot to opt out and I won't know who you are

I will make every bit of data as easy and accessible for people to train on and use, I'm worried about practically helping the ecosystem, not semantics. If you have a practical problem - e.g. removing your data - I'm happy to help. If not and you care enough to complain, I'd suggest you put that energy into doing something useful for open source AI

And if you have a practical issue with this, please provide your username and I'll make sure you're removed Though that same data will undoubtedly be sold by Discord and others who will not provide you with a way to opt out, such is the way of the world