iiinova — where your digital self keeps evolving

Reposted by Lain

merve @merve.bsky.social · 26/11/2024

Small yet mighty! 💫 We are releasing SmolVLM: a new 2B small vision language made for on-device use, fine-tunable on consumer GPU, immensely memory efficient 🤠 We release three checkpoints under Apache 2.0: SmolVLM-Instruct, SmolVLM-Synthetic and SmolVLM-Base huggingface.co/collections/...

1115927

Lain @not-so-lain.bsky.social · 25/11/2024

new loadimg release with more robust url handling here's an example using fal as an API

020

Lain @not-so-lain.bsky.social · 25/11/2024

Found this little easter egg in one of the blogposts on the HUB. Apart from efficient upload/download of lfs files this would be my second favorite feature that xet team is working on. My expectations for the xethub team is off the roof 🔥

030

Lain @not-so-lain.bsky.social · 23/11/2024

I have a question about dataset collection, do you manually annotate all of them ? or do you look for good ones out there and directly use them?

000

Reposted by Lain

Xenova @xenova.bsky.social · 14/11/2024

Introducing Transformers.js v3: State-of-the-art Machine Learning for the Web. New features include: ⚡ WebGPU support (up to 100x faster than WASM) 🔢 New quantization formats 🏛 121 supported architectures in total 🤖 Over 1200 pre-converted models Get started with `npm i @huggingface/transformers`

4337

Lain @not-so-lain.bsky.social · 14/11/2024

Did you know you can make API requests to VQA models on huggingface by sending the base64 images in the url field 👀 this allows you to work with local images without the need to upload them online first 🔥 simplified code in the link below huggingface.co/posts/not-la...

huggingface.co

@not-lain on Hugging Face: "ever wondered how you can make an API call to a visual-question-answering…"

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

000

Lain @not-so-lain.bsky.social · 13/11/2024

awesome, thanks a lot for sharing 🙌

011

Reposted by Lain

Aritra Roy Gosthipaty @arig23498.bsky.social · 13/11/2024

`bitsandbytes` makes it really easy to quantize models Note: MB should be GB in the diagram.

171

Lain @not-so-lain.bsky.social · 13/11/2024

cool, looking forward to the next release 🙌

000

Lain @not-so-lain.bsky.social · 13/11/2024

new SOTA for background removal beats birefnet 🚀 api ready for personal use 👀 🔗 huggingface.co/spaces/briaa...

010

Lain @not-so-lain.bsky.social · 13/11/2024

are you going to teach moondream latex ?

110

Lain @not-so-lain.bsky.social · 13/11/2024

oh thanks a lot for the reply, really awesome work tho ✨

000

Lain @not-so-lain.bsky.social · 12/11/2024

didn't know molmo supports multiple image visual question answering, any tips or papers you might share with me ?

100

Lain @not-so-lain.bsky.social · 11/11/2024

got invited to yet another AI event

000

Lain @not-so-lain.bsky.social · 11/11/2024

published a blogpost about integrating custom AI models with huggingface leveraging PyTorchModelHubMixin. highlights : * automodel * custom input types * trainer api support * etc , .... Hope this is informative 🤗 huggingface.co/blog/not-lai...

huggingface.co

PyTorchModelHubMixin: Bridging the Gap for Custom AI Models on Hugging Face

A Blog post by Lain on Hugging Face

020

Lain @not-so-lain.bsky.social · 01/11/2024

how come sasha does not have a 🦋 account ?

100

Reposted by Lain

merve @merve.bsky.social · 31/10/2024

Another great week in open ML! Here's a small thread with this week's recap 🧵 💬 Small language models @huggingface released SmolLM2, a new family of smol language models with Apache 2.0 license with, along with datasets of sizes 135M, 360M and 1.7B 🤗

2299

Lain @not-so-lain.bsky.social · 31/10/2024

First post on this app Happy Halloween (≧∇≦)ﾉ✨

041