CTO @ https://xitter.twitterapi.xyz/t.co/xeM0KpR9AC - Free opensource AI Video production. warpfusion, ArcaneGAN, face2comics. All tweets are sarcastic unless stated otherwise.
Joined December 2011
- Tweets2.7K
- Following1.4K
- Followers7.3K
- Likes14.2K
Pinned Tweet
We're living up to the promise (unlike others, wink wink) and open-sourcing Mago Studio today
Use you local GPU via @ComfyUI , or a @vast_ai / @runpod instance.
Prompt a chatbot to use your preferred model weights or workflows - no problem, update anything to your taste - it just works, because there is already a solid foundation based on the time-tested MagoStudio webapp.
LMK what features or models you'd like to see in the local version next in the comments
The code is here - github.com/MagoStudio/MagoSt…
Win, Max, Ubuntu builds (if you can't run npm run dev :D) - github.com/MagoStudio/MagoSt…
Ok it looks like we don't really need a standalone minimax H3 image 20gb checkpoint :D
ref2va + @pika_labs flex gets similar results (6s vs 3s and same vram usage) at 1024px
2k and 4k results below 👇
4K image generation in 10 seconds! ⚡
Meet MiniMax H3-Image.
16 million pixel resolution. Stunning realism. Precise prompt following & editing. Up to 16 reference images. All in 10 seconds.
And it's open-weight. 🚀
Available now via our API, MiniMax Code, and Hailuo AI — with 50% off 16MP+ generations during launch week!
Can't wait to see what you create. 🔥
Hugging Face: huggingface.co/MiniMaxAI/Min…
API: platform.minimax.io/docs/gui…
MiniMax Code: code.minimax.io/about
congratz! what a time to be alive!
4K image generation in 10 seconds! ⚡
Meet MiniMax H3-Image.
16 million pixel resolution. Stunning realism. Precise prompt following & editing. Up to 16 reference images. All in 10 seconds.
And it's open-weight. 🚀
Available now via our API, MiniMax Code, and Hailuo AI — with 50% off 16MP+ generations during launch week!
Can't wait to see what you create. 🔥
Hugging Face: huggingface.co/MiniMaxAI/Min…
API: platform.minimax.io/docs/gui…
MiniMax Code: code.minimax.io/about
8
577
Well with Qwen3.8 on 5090 at 300 t/s you can write 25m x 0.6$ = 15$ per day. So you'd save 5700$ per year minus electricity.
And it's low GPU load sice it's mostly memory bandwidth.
Ram Altman pls log in
2
6
786
Ok here we go!
Disco WarpFusion v0.9.1:
Turbo and quality loras support
6-step turbo lora (vs 30 step vanilla - 5x speedup) and aesthetic quality lora (both early stage previews)
Optional Inference Optimizations
- compile CLIP and secondary
- batch CLIP cuts
- lower precision
- lora merge
about +30% speedup with all enabled
and other tweaks + stuff from stable warp notebook, more details here - patreon.com/sxela/posts/disc… (public)
Feeling cute, might release Disco WarpFusion v0.9.1 notebook tonight
time to unstash your settings.txts!
Added multilingual support to local MagoStudio
You can now add your languages using the pr's template: github.com/MagoStudio/MagoSt…
We didn't want to vibe-translate since the context matters, especially for this specific niche :D
1
4
438
the best thing ofc is that you can cut the render times on 40% with all the modern bells and whistles in just one codex prompt, and then leave a parameter sweep overnight to find the best settings
Added a pr with @pika_labs Minimax h3 flex loras by @_akhaliq
Show some love by likes or comments to the issue so the product team lets me merge it 😅🫶
github.com/MagoStudio/MagoSt…
1
2
297