Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Omar Sanseviero Profile picture
Omar Sanseviero
@osanseviero
Making ML go brr at Google ex-Chief Llama Officer @huggingface 🦙 Founder @AI_Learners. 100% Hacker Llama🇵🇪🇲🇽
2 subscribers
Dec 24 • 4 tweets • 2 min read
Introducing a holidays pet project: GeminiCoder! Go wild and create apps in seconds just with a prompt 🚀



Share your creations! 👀hf.co/spaces/osansev… This project is a fork from @nutlope's great github.com/nutlope/llamac…

Find the used code at github.com/osanseviero/ge…
May 19 • 10 tweets • 3 min read
What a week in the ML world! Here is a recap thread on all the exciting open ML updates!🔥

VLMs: Salesforce, Kosmos 2.5, PaliGemma, Cumo
LLM: Yi 1.5, Falcon 2, DeepSeek v2 lite
Diffusion: HunyuanDiT, Lumina next

Keep reading 👇 1. Salesforce released BLIP3 (renamed to xGen-mm), a visual LLM family

May 8, 2023 • 5 tweets • 4 min read
Open Source ML is going brrr 🚀 Here are 5 amazing free OS demos

1⃣Stable Diffusion + ControlNet for hand control

Demo: hf.co/spaces/vllab/c…
Look at those fingers! 😍

⬅️+promp = Input
➡️ = Output ImageImage 2⃣Text-to-Motion Retrieval 🪅

Demo hf.co/spaces/Mathux/…
Very cool work from @Michael_J_Black @gulvarol and team!

Input: Someone is jumping with one foot
Output:
May 4, 2023 • 5 tweets • 4 min read
💫StarCoder: May the Source be with You!

🔥15B LLM with 8k context
🥳Trained on permissively-licensed code
💻Acts as tech assistant
🤯80+ programming languages
🚀Open source and data
💫Online demos
🧑‍💻VSCode plugin
🪅1 trillion tokens

Follow this amazing experience! 🧵 Image The model can generate and complete code, but it also has nice chatty properties. Some things it can do

- Write code snippets
- Translate Python to C++
- Explain concepts (what's recursion)
- Act as a terminal
- Tell me how to exit vim
🔥 Image
Jan 15, 2023 • 10 tweets • 4 min read
Graphs are **everywhere**, from social media and knowledge systems to molecules and meshes! 🧑‍🏫

Want to learn about Machine Learning for Graphs? Check out this thread! 🧵 Graphs are denoted as G(V,E), with V being its vertices (nodes in the graph) and E being its edges (connections or links).

In a social network, the vertices would be its users, and the edges would be the connections. 🐦
Dec 9, 2022 • 5 tweets • 2 min read
🧵I asked ChatGPT to explain what Transfer Learning is, but
- Rapping
- As if I were 5 years old
- A corn lover
- To Homer Simpson teaching Bart
- As Schmidhuber talking with Geoffrey Hinton

Follow this thread! Transfer learning but with some rap! 🎶 Image Using corn analogies 🌽 Image
Nov 25, 2022 • 6 tweets • 3 min read
🧵Some cool ML demos loved by the community🔥

[1/6] Multilingual Stable Diffusion

huggingface.co/spaces/juancop…

Stable Diffusion Pipeline that supports prompts in 50 different languages. 🌎🌍🌏 [2/6] PlaygroundAI Images Explorer 🔍

Explore, inspect and analyze thousands of images from PlaygroundAI, with its prompt, source, and more!

huggingface.co/spaces/ysharma…
Nov 20, 2022 • 13 tweets • 5 min read
🧵 Some thoughts about the recent release of Galactica by @MetaAI (everything here is my personal opinion) 👀

Let's start with the positive / What went well [1] The model was released and Open Source*

Contrary to the trend of very interesting research being closed or just accessible through paid APIs, by open-sourcing the models and building on top of existing OS tools, evaluation can be reliably done in a transparent and open way
Oct 31, 2022 • 8 tweets • 5 min read
This week @huggingface Spaces of the week are 🔥Check this thread for their respective announcements, or go directly and play with the demos!

🧵Check them out! huggingface.co/spaces Image [1/7] Stable Diffusion Multiplayer by @radamar is an amazing collaborative experiment in which hundreds of people interact to paint a canvas using Stable Diffusion

Oct 26, 2022 • 14 tweets • 6 min read
CLIP Interrogator + Stable Diffusion

I used CLIP Interrogator to generate the prompt of an image, then I fed the prompt to SD, and used the image in CLIP Interrogator...and again!

🧵Follow along in this llamastic journey! Image before / after ImageImage
Oct 24, 2022 • 6 tweets • 2 min read
I just published a new edition of the free weekly ML News 🔥

Check it out! 👀
thehackerllama.substack.com/p/machine-lear… Three interesting papers about editing images with text (Imagic, DiffEdit, and Prompt to Prompt)

Oct 11, 2022 • 6 tweets • 6 min read
Six open-source ML demos from the last 6 days 🔥

1. Stable Diffusion Infinity 🎨

Outpaint Stable Diffusion on an infinite canvas

Demo huggingface.co/spaces/lnyan/s…
GitHub github.com/lkwq007/stable… by @lkwq007 2. Positive Reframing 🤗

Start inputting negative texts to see how you can see the same event from a positive angle.

Choose among different strategies and reframe them to behave more positively! 😀

huggingface.co/spaces/Ella232…
Sep 26, 2022 • 8 tweets • 6 min read
🧵[1/8] Did you know you can find cool free top demos at hf.co/spaces (Spaces of the week)?

Let's explore them!

🔊@OpenAI Whisper Multilingual Speech Recognition model demo allows you to speak and get a transcript in seconds! [2/8] Stable Diffusion Conceptualizer 🧑‍🎨

Navigate 500+ community-taught objects and styles and create images using them!

huggingface.co/spaces/sd-conc…
Sep 5, 2022 • 4 tweets • 3 min read
Did you know you can check out the trending ML demos at hf.co (at the right in trending)

Check out some of them🧵👇 1. Stable Diffusion huggingface.co/spaces/stabili…
2. Diffuse the Rest - img2img app huggingface.co/spaces/hugging…
3. ERNIE-ViLG - sota Chinese text to image huggingface.co/spaces/PaddleP…
Aug 22, 2022 • 5 tweets • 3 min read
🧵Stable Diffusion weights are officially public, and we got some surprises! 🤗

🤗 Public weights
🧨Support with the diffusers library
🔥Load and use the model with a few lines of code
📖Blog post explaining how it works

github.com/huggingface/di… Image We wrote a blog post explaining how to control the pipeline and an overview of how Stable Diffusion works.

Check it out👉 hf.co/blog/stable_di… Image
Aug 7, 2022 • 12 tweets • 6 min read
🧵Train a @fastdotai model to classify llamas vs alpacas, build a nice interactive demo and share it with your friends in <10 minutes. See how in this thread!

Demo: hf.co/spaces/hugging… We begin by creating a dataset. Here we use DuckDuckGo to download ~200 images for llamas and 200 images from alpacas. 🦙

The images are saved in a directory named with the label (i.e. llamas are saved in /llama directory)
Aug 3, 2022 • 5 tweets • 4 min read
🧵[1/5] Let's take a look at some interesting open-source ML demos!

SegFormer implemented in @TensorFlow and converted to ONNX for scene segmentation.⚔️

hf.co/spaces/chansun… [2/5] Multi Task NLP 🔥

Learn about different NLP tasks by exploring this interactive demo from keyword extraction to emotion detection. Very cool!

huggingface.co/spaces/miesner…
Jul 5, 2022 • 8 tweets • 3 min read
🧵Many people asking how they can contribute to Open Source, here is a thread of different ways to do so! ⤵️ [1] 📓Not all contributions are code. One of the best ways to get started is by contributing documentation, from small fixes to adding new docstrings, everything is helpful!

Google has a neat course for technical writing you can check at developers.google.com/tech-writing
Jun 30, 2022 • 5 tweets • 2 min read
🧵[1/5] A month ago we launched community features in @huggingface Hub. Since then, the team has been working hard in adding lots of new community features. Let's go through them! 🤗

You can now react to discussions with cool emojis! [2/5] you can create draft Pull Requests 🥳🥳🎉🦙
Jun 26, 2022 • 11 tweets • 4 min read
🧵[1/n] Sentence Embeddings can be extremely powerful and useful for many applications. Learn how to build a FAQ system in the following blog!

huggingface.co/blog/getting-s… [2/n] Embeddings represent a sentence in a vector space in which its semantic meaning is captured.

For example, "What is the main benefit of voting"
can be represented as a list of 384 numbers
[0.8, 0.4, ..., 0.7] 🧑‍🔬
Jun 13, 2022 • 5 tweets • 3 min read
🧵[1/5] 5 Cool ML demos

Talking Face Multilingual Text to Speech
📝Write a text
🌍Translate
🔊Generate a video with a head saying what you wrote

hf.co/spaces/CVPR/ml…

See "Hello, this is demonstration for talking face generation with multilingual text-to-speech" in Korean 🔥 [2/5] It Happened One Frame 📹

Search for a frame in a YouTube video given a description you give. It works quite nicely!

hf.co/spaces/YiYiXu/…