Hugging Face's logo

Huggingface.js
community
https://github.com/huggingface/huggingface.js
Activity Feed

AI & ML interests

Collection of JS libraries to interact with the Hugging Face Hub

Recent Activity

coyotte508Β  updated a Space 16 days ago
huggingfacejs/better-auth-test
coyotte508Β  published a Space 16 days ago
huggingfacejs/better-auth-test
coyotte508Β  new activity about 1 month ago
huggingfacejs/client-side-oauth:Update index.html
View all activity

merveΒ 
posted an update 3 days ago
view post
Post
578
SOOOO MANY MODEL RELEASES 😍
Here's some picks from past week πŸ€—

> ByteDance/XVerse is a new identity preserving image generation model πŸ–ΌοΈ
> google/gemma-3n-E4B-it, any-to-text model supported by transformers πŸ€—
> nvidia/llama-nemoretriever-colembed-3b-v1 two new state-of-the-art visual document retrievers πŸ“‘
> New version of Dia TTS model is up nari-labs/Dia-1.6B-0626
> Black Forest Labs releases Kontext benchmark black-forest-labs/kontext-bench

Find more here merve/releases-june-27-6864e8eb17f7e3a8b444083c
merveΒ 
posted an update 3 days ago
merveΒ 
posted an update 5 days ago
merveΒ 
posted an update 9 days ago
view post
Post
560
Dataset Viewer for PDFs just landed on Hugging Face πŸ“–πŸ€— you can now preview all the PDFs easier than before!

on top of this, there's PdfFolder format to load the PDF datasets quicker πŸ’¨
> to use it, your dataset should follow a directory format like folder/train/doc1.pdf, folder/train/doc1.pdf
> if you want to include bounding boxes, labels etc. you can keep them in a metadata.csv file in the same folder 🀝

read document dataset docs https://huggingface.co/docs/datasets/main/en/document_dataset
check all the document datasets here https://huggingface.co/datasets?modality=modality:document&sort=trending πŸ“–
  • 1 reply
Β·
merveΒ 
posted an update 11 days ago
view post
Post
612
we've merged LightGlue keypoint matcher to Hugging Face transformers! it allows commercial use when paired with an open-source keypoint detector πŸ™πŸ»

it works very well, try it yourself: ETH-CVG/LightGlue

here's an in-the-wild test with two images of the same place ‡️
  • 1 reply
Β·
merveΒ 
posted an update 12 days ago
view post
Post
4307
Release picks of the past week is here! Find more models, datasets, Spaces here merve/june-20-releases-68594824d1f4dfa61aee3433

πŸ–ΌοΈ VLMs/OCR
> moonshotai/Kimi-VL-A3B-Thinking-2506 is a powerful reasoning vision LM, 3B active params, smarter with less tokens, supports long documents, videos πŸ‘ (OS)
> nanonets/Nanonets-OCR-s is 3.75B params OCR model based on Qwen2.5VL-3B-Instruct (OS)

πŸ’¬ LLMs
> moonshotai/Kimi-Dev-72B is a strong coding model based on Qwen2.5-72B (OS)
> Mistral released mistralai/Mistral-Small-3.2-24B-Instruct-2506, an update to their former model with better function calling & instruction following (OS)

πŸ—£οΈ Audio
> Google released google/magenta-realtime, real time music generation & audio synthesis (cc-by-4)
> kyutai released new speech-to-text models that come in 1B & 2B ( kyutai/stt-1b-en_fr, stt-2b-en_fr) with 0.5s and 2.5s delay

3D
> Tencent released tencent/Hunyuan3D-2.1 an image-to-3D model (see below)
merveΒ 
posted an update 13 days ago
merveΒ 
posted an update 15 days ago
coyotte508Β 
updated a Space 16 days ago
coyotte508Β 
published a Space 16 days ago
merveΒ 
posted an update 16 days ago
view post
Post
1903
stop using VLMs blindly βœ‹πŸ»

compare different VLM outputs on a huge variety of inputs (from reasoning to OCR!) πŸ”₯ visionLMsftw/comparevlms

> has support for multiple VLMs: google/gemma-3-27b-it, Qwen/Qwen2.5-VL-7B-Instruct, Qwen/Qwen2.5-VL-32B-Instruct, meta-llama/Llama-4-Maverick-17B-128E-Instruct, HuggingFaceTB/SmolVLM2-2.2B-Instruct
> recommend us new models or inputs, we'll add 🫑

so far I figured out
> for fact-checks, you need a relatively bigger size (7B is ok!)
> Gemma 3 gets downgrade without pan and scan (especially for πŸ“‘)
> Qwen2.5VL-32B is very talkative, great for reasoning but not good for simple tasks πŸ—£οΈ
  • 2 replies
Β·
merveΒ 
posted an update 17 days ago
view post
Post
3606
Releases of the past week are here merve/releases-june-13-6852c3c1eaf1e0c24c958860

Here's our picks πŸ€“
So many interesting models released past week in open AI! πŸ€–

πŸ–ΌοΈ Computer Vision/VLMs
> nanonets/Nanonets-OCR-s is the new state-of-the-art OCR model that can handle checkboxes, watermarks, tables (OS)
> Meta released facebook/v-jepa-2-6841bad8413014e185b497a6, new sota video embeddings with two new classification models (OS)
> ByteDance-Seed/SeedVR2-3B is a new 3B video restoration model (OS)

Audio
> Stepfun released stepfun-ai/Step-Audio-AQAA, new large (137B 🀯) audio language model that takes in audio and generates audio (OS)

πŸ€– Robotics
> nvidia released nvidia/GR00T-N1.5-3B, new open foundation vision language action model

3D
> tencent/Hunyuan3D-2.1 is the new version of Hunyuan by Tencent that can generate 3D assets from text and image prompts
merveΒ 
posted an update 18 days ago
view post
Post
3542
IN: video fine-tuning support for facebook V-JEPA 2 in HF transformers πŸ”₯

it comes with
> four models fine-tuned on Diving48 and SSv2 dataset facebook/v-jepa-2-6841bad8413014e185b497a6
> FastRTC demo on V-JEPA2 SSv2 qubvel-hf/vjepa2-streaming-video-classification
> fine-tuning script on UCF-101 https://gist.github.com/ariG23498/28bccc737c11d1692f6d0ad2a0d7cddb
> fine-tuning notebook on UCF-101 https://colab.research.google.com/drive/16NWUReXTJBRhsN3umqznX4yoZt2I7VGc?usp=sharing
we're looking forward to see what you will build! πŸ€—
merveΒ 
posted an update 19 days ago
view post
Post
2445
#CVPR2025 Paper Picks #1
VisionZip is a compression technique that reduces number of visual tokens to improve performance AND prefill time for vision language models
demo: Senqiao/VisionZip
paper: VisionZip: Longer is Better but Not Necessary in Vision Language Models (2412.04467)
most of the image tokens are redundant for the LLM, so the authors ask "are all visual tokens necessary?"

the method is simple:
find which tokens have the highest attention score, merge rest of the tokens based on similarity, then merge both

their method is both training-free and for fine-tuning
the authors report 5 point improvement on average of vision language tasks + 8x improvement in prefilling time for Llava-Next 7B and 13B 🀯

removing redundant tokens improve image token quality too πŸ₯Ή
merveΒ 
posted an update 19 days ago
view post
Post
3678
stop writing CUDA kernels yourself

we have launched Kernel Hub: easy optimized kernels for all models on Hugging Face πŸ”₯ use them right away!
it's where the community populates optimized kernels 🀝

this release comes in three parts
> Kernel Hub: contains (as of now) 14 kernels
> kernels: Python library to load kernels from Kernel Hub
> kernel-builder: Nix package to build kernels for PyTorch (made using PyTorch C++ frontend)

when building models, your regular workflow should be pulling kernels from Hub and building your model with them πŸ€—
here's a practical example with RMSNorm:
1. pull the kernel from Hub with get_kernel
2. decorate with use_kernel_forward_from_hub
3. inject it to your model
we'd love to hear your feedback! πŸ™πŸ»
we also welcome kernel contributions by community πŸ₯ΉπŸ’—

- request kernels here: kernels-community/README#1
- check out this org: kernels-community
- read the blog: https://huggingface.co/blog/hello-hf-kernels
  • 1 reply
Β·
merveΒ 
posted an update 22 days ago
view post
Post
719
Dolphin: new OCR model by ByteDance with MIT license 🐬

the model first detects element in the layout (table, formula etc) and then parses each element in parallel for generation
Model: ByteDance/Dolphin
Try the demo: ByteDance/Dolphin
merveΒ 
posted an update 24 days ago
view post
Post
1377
stop building parser pipelines πŸ‘‹πŸ»
there's a new document parser that is small, fast, Apache 2.0 licensed and is better than all the other ones! 😱

echo840/MonkeyOCR is a 3B model that can parse everything (charts, formules, tables etc) in a document 🀠
> the authors show in the paper that document parsing pipelines often have errors propagating back
> using singular e2e models are better but they're too heavy to use

this model addresses both: it's lighter, faster, stronger πŸ”₯
merveΒ 
posted an update 24 days ago
view post
Post
1607
Meta just released V-JEPA 2: new open-source image/video world models β―οΈπŸ€— facebook/v-jepa-2-6841bad8413014e185b497a6

> based on ViT, different sizes (L/G/H) and resolution (286/384)
> 0-day support in πŸ€— transformers
> comes with a physical reasoning (from video) benchmark: MVPBench, IntPhys 2, and CausalVQA facebook/physical_reasoning_leaderboard

Read more https://ai.meta.com/blog/v-jepa-2-world-model-benchmarks/
We will release a fine-tuning notebook with task-specific models in transformers format soon, stay tuned!
merveΒ 
posted an update 30 days ago
XenovaΒ 
posted an update about 1 month ago
view post
Post
5032
NEW: Real-time conversational AI models can now run 100% locally in your browser! 🀯

πŸ” Privacy by design (no data leaves your device)
πŸ’° Completely free... forever
πŸ“¦ Zero installation required, just visit a website
⚑️ Blazingly-fast WebGPU-accelerated inference

Try it out: webml-community/conversational-webgpu

For those interested, here's how it works:
- Silero VAD for voice activity detection
- Whisper for speech recognition
- SmolLM2-1.7B for text generation
- Kokoro for text to speech

Powered by Transformers.js and ONNX Runtime Web! πŸ€— I hope you like it!
Β·