Hugging Apps

Tweeting the coolest apps and demos on @huggingface Spaces
curation by @multimodalart

[Will Lin @wlsaidhi]

FastVideo's @MiniMax_AI FastH3 Omni model is now available on @huggingface spaces! Try it! Up to 9 reference images, 3 videos, and 3 audio clips!

Thanks to @multimodalart for adding support!

https://t.co/Ul46IHw3Ov

[Hugging Apps @HuggingApps]

A 4B vision model that points at what it's talking about 🕶️

Xiaomi-Lens-0 answers in plain language, and every object it mentions comes with its own segmentation mask.

https://t.co/CokWtfiAzH

[Hugging Apps @HuggingApps]

DuET: piano audio → sheet music 🎹

DuET transcribes a piano recording straight into an engraved score: notes, chords, key and time signature 🎵

🎶 https://t.co/wynmFccLBz

[apolinario (poli) @multimodalart]

yay! FastH3 Omni is now open weights! 📹

high quality 8 step reference-to-video, fully end-to-end open source (including training code)

this is the model many said rivals fal's H3 Max 👀

▶️ on Spaces https://t.co/C3eQBwQ0LN

📁 Model https://t.co/gKFzE2r5mv

[Hugging Apps @HuggingApps]

Shoot first, focus later 📷

FlashReFocus restores a photo focus for the entire image, then refocuses wherever you click, in ~0.6 s per click.

▶️ https://t.co/6va3Aq4pZJ

[Hugging Apps @HuggingApps]

BRIA dropped FIBO 2.0 Edit Turbo on @huggingface edits images precisely in only 4 steps ⚡

One instruction, converted into a precise json, up to 4 reference images, or a mask to change focus in one area

▶️ on Spaces
https://t.co/kjhBpjGmV7

[Hugging Apps @HuggingApps]

SemBridge is here: a model that speaks and sings in the same voice 🎤🔊

https://t.co/tqqMpN9kP6

[Hugging Apps @HuggingApps]

FLUX.3 Image Klein 6B just dropped on @huggingface 🔥

Base and distilled (16 steps), multi-reference editing, JSON bounding-box layouts

play with it 👇
https://t.co/QedfRJm5IB

[Linoy Tsaban @linoy_tsaban]

ICYMI: Lightx2v just dropped Qwen Image 2.1 Lightning ⚡️

> 4-step distilled LoRA
> weights: https://t.co/QPXZgrfwvL
> space: https://t.co/K9Sx7MZyw5

[Linoy Tsaban @linoy_tsaban]

New H3 character swap LoRA🔥
amazing work from @Alissonerdx

LoRA: https://t.co/ggpfdCBlST
Space:
https://t.co/81Ub3URxrm

[Hugging Apps @HuggingApps]

one photo, new camera angles 📷

this Qwen-Image-2.1 LoRA by @_akhaliq using the ML Intern reshoots a subject from a camera position you pick: close-up, high angle, straight down, from behind. no background

▶️ on Spaces https://t.co/RRchljjoTS

[Hugging Apps @HuggingApps]

@frendudebro @huggingface thanks for the feedback. I had good experiences with the quality in the initial tests, but the reason we build apps is so people can try for themselves

[Hugging Apps @HuggingApps]

Qwen Image 2.1 Turbo just landed on @huggingface
8 steps, same quality

▶️ on Spaces https://t.co/3cmXihIAYJ

[Hugging Apps @HuggingApps]

pose a mannequin, anyone takes the pose 🕺

the VNCCS PoseStudio LoRA for Qwen-Image-2.1 redraws any character in the exact pose of a 3D mannequin, keeping their face, clothes and art style

▶️ on Spaces https://t.co/QNMr5PSRVp

[Hugging Apps @HuggingApps]

@whataboutfranks fixed!

[Hugging Apps @HuggingApps]

native 4K video, from a 1.3B model 🔍

T3-Video retrofits Wan2.1-1.3B with window attention so it generates directly in 3840×2176, no upscaler: zoom in and every bush is still there

▶️ on Spaces https://t.co/56ijVxy3JF

[Hugging Apps @HuggingApps]

⬇️
https://t.co/afSMhE5itG

[Sahil Mahendrakar @sahilmdkr]

@HuggingApps hey thanks for making this space! if you're interested in learning more about how I distilled Kokoro I have an interactive blog post here https://t.co/VjUeROL7DT and a paper here https://t.co/TPbzkXklPj

[Hugging Apps @HuggingApps]

Kokoro TTS, but 10× smaller 🫰

Paradee distills Kokoro-82M into a 8M-param TTS

just 9 MB of weights can speak faster than real time on a single CPU thread and runs on your toaster

▶️ on Spaces https://t.co/0Pj4FdC2wO

[Linoy Tsaban @linoy_tsaban]

Doing video background replacement with H3 Inpainting + LTX Alpha Matte LoRA, and it's surprisingly good 🔥

> LTX 2.5 LoRA to extract matte
> H3 Ref inpainting (+ Turbo LoRA) to generate a new background
> Mask drop knob to reduce flickering and improve lighting match - drops the inpainting mask at step t

I added the workflow to 🧨diffusers modular + built Spaces for it 👇