Hugging Quants

Activity Feed

AI & ML interests

Optimised quants for high-throughput deployments! Compatible with Transformers, TGI & vLLM ๐Ÿค—

Recent Activity

hugging-quants's activity

Xenovaย 
posted an update 4 days ago
view post
Post
5054
We did it. Kokoro TTS (v1.0) can now run 100% locally in your browser w/ WebGPU acceleration. Real-time text-to-speech without a server. โšก๏ธ

Generate 10 seconds of speech in ~1 second for $0.

What will you build? ๐Ÿ”ฅ
webml-community/kokoro-webgpu

The most difficult part was getting the model running in the first place, but the next steps are simple:
โœ‚๏ธ Implement sentence splitting, allowing for streamed responses
๐ŸŒ Multilingual support (only phonemization left)

Who wants to help?
ยท
Xenovaย 
posted an update 26 days ago
view post
Post
5227
Introducing Kokoro.js, a new JavaScript library for running Kokoro TTS, an 82 million parameter text-to-speech model, 100% locally in the browser w/ WASM. Powered by ๐Ÿค— Transformers.js. WebGPU support coming soon!
๐Ÿ‘‰ npm i kokoro-js ๐Ÿ‘ˆ

Try it out yourself: webml-community/kokoro-web
Link to models/samples: onnx-community/Kokoro-82M-ONNX

You can get started in just a few lines of code!
import { KokoroTTS } from "kokoro-js";

const tts = await KokoroTTS.from_pretrained(
  "onnx-community/Kokoro-82M-ONNX",
  { dtype: "q8" }, // fp32, fp16, q8, q4, q4f16
);

const text = "Life is like a box of chocolates. You never know what you're gonna get.";
const audio = await tts.generate(text,
  { voice: "af_sky" }, // See `tts.list_voices()`
);
audio.save("audio.wav");

Huge kudos to the Kokoro TTS community, especially taylorchu for the ONNX exports and Hexgrad for the amazing project! None of this would be possible without you all! ๐Ÿค—

The model is also extremely resilient to quantization. The smallest variant is only 86 MB in size (down from the original 326 MB), with no noticeable difference in audio quality! ๐Ÿคฏ
ยท
Xenovaย 
posted an update about 1 month ago
view post
Post
8267
First project of 2025: Vision Transformer Explorer

I built a web app to interactively explore the self-attention maps produced by ViTs. This explains what the model is focusing on when making predictions, and provides insights into its inner workings! ๐Ÿคฏ

Try it out yourself! ๐Ÿ‘‡
webml-community/attention-visualization

Source code: https://github.com/huggingface/transformers.js-examples/tree/main/attention-visualization
Xenovaย 
posted an update about 2 months ago
view post
Post
4110
Introducing Moonshine Web: real-time speech recognition running 100% locally in your browser!
๐Ÿš€ Faster and more accurate than Whisper
๐Ÿ”’ Privacy-focused (no data leaves your device)
โšก๏ธ WebGPU accelerated (w/ WASM fallback)
๐Ÿ”ฅ Powered by ONNX Runtime Web and Transformers.js

Demo: webml-community/moonshine-web
Source code: https://github.com/huggingface/transformers.js-examples/tree/main/moonshine-web
ยท
Xenovaย 
posted an update 2 months ago
view post
Post
3199
Introducing TTS WebGPU: The first ever text-to-speech web app built with WebGPU acceleration! ๐Ÿ”ฅ High-quality and natural speech generation that runs 100% locally in your browser, powered by OuteTTS and Transformers.js. ๐Ÿค— Try it out yourself!

Demo: webml-community/text-to-speech-webgpu
Source code: https://github.com/huggingface/transformers.js-examples/tree/main/text-to-speech-webgpu
Model: onnx-community/OuteTTS-0.2-500M (ONNX), OuteAI/OuteTTS-0.2-500M (PyTorch)
reach-vbย 
posted an update 2 months ago
view post
Post
4876
VLMs are going through quite an open revolution AND on-device friendly sizes:

1. Google DeepMind w/ PaliGemma2 - 3B, 10B & 28B: google/paligemma-2-release-67500e1e1dbfdd4dee27ba48

2. OpenGVLabs w/ InternVL 2.5 - 1B, 2B, 4B, 8B, 26B, 38B & 78B: https://huggingface.co/collections/OpenGVLab/internvl-25-673e1019b66e2218f68d7c1c

3. Qwen w/ Qwen 2 VL - 2B, 7B & 72B: Qwen/qwen2-vl-66cee7455501d7126940800d

4. Microsoft w/ FlorenceVL - 3B & 8B: https://huggingface.co/jiuhai

5. Moondream2 w/ 0.5B: https://huggingface.co/vikhyatk/

What a time to be alive! ๐Ÿ”ฅ
dvilasueroย 
posted an update 2 months ago
view post
Post
2360
๐ŸŒ Announcing Global-MMLU: an improved MMLU Open dataset with evaluation coverage across 42 languages, built with Argilla and the Hugging Face community.

Global-MMLU is the result of months of work with the goal of advancing Multilingual LLM evaluation. It's been an amazing open science effort with collaborators from Cohere For AI, Mila - Quebec Artificial Intelligence Institute, EPFL, Massachusetts Institute of Technology, AI Singapore, National University of Singapore, KAIST, Instituto Superior Tรฉcnico, Carnegie Mellon University, CONICET, and University of Buenos Aires.

๐Ÿท๏ธ +200 contributors used Argilla MMLU questions where regional, dialect, or cultural knowledge was required to answer correctly. 85% of the questions required Western-centric knowledge!

Thanks to this annotation process, the open dataset contains two subsets:

1. ๐Ÿ—ฝ Culturally Agnostic: no specific regional, cultural knowledge is required.
2. โš–๏ธ Culturally Sensitive: requires dialect, cultural knowledge or geographic knowledge to answer correctly.

Moreover, we provide high quality translations of 25 out of 42 languages, thanks again to the community and professional annotators leveraging Argilla on the Hub.

I hope this will ensure a better understanding of the limitations and challenges for making open AI useful for many languages.

Dataset: CohereForAI/Global-MMLU
Xenovaย 
posted an update 3 months ago
view post
Post
4073
We just released Transformers.js v3.1 and you're not going to believe what's now possible in the browser w/ WebGPU! ๐Ÿคฏ Let's take a look:
๐Ÿ”€ Janus from Deepseek for unified multimodal understanding and generation (Text-to-Image and Image-Text-to-Text)
๐Ÿ‘๏ธ Qwen2-VL from Qwen for dynamic-resolution image understanding
๐Ÿ”ข JinaCLIP from Jina AI for general-purpose multilingual multimodal embeddings
๐ŸŒ‹ LLaVA-OneVision from ByteDance for Image-Text-to-Text generation
๐Ÿคธโ€โ™€๏ธ ViTPose for pose estimation
๐Ÿ“„ MGP-STR for optical character recognition (OCR)
๐Ÿ“ˆ PatchTST & PatchTSMixer for time series forecasting

That's right, everything running 100% locally in your browser (no data sent to a server)! ๐Ÿ”ฅ Huge for privacy!

Check out the release notes for more information. ๐Ÿ‘‡
https://github.com/huggingface/transformers.js/releases/tag/3.1.0

Demo link (+ source code): webml-community/Janus-1.3B-WebGPU
reach-vbย 
posted an update 3 months ago
view post
Post
4595
Massive week for Open AI/ ML:

Mistral Pixtral & Instruct Large - ~123B, 128K context, multilingual, json + function calling & open weights
mistralai/Pixtral-Large-Instruct-2411
mistralai/Mistral-Large-Instruct-2411

Allen AI Tรผlu 70B & 8B - competive with claude 3.5 haiku, beats all major open models like llama 3.1 70B, qwen 2.5 and nemotron
allenai/tulu-3-models-673b8e0dc3512e30e7dc54f5
allenai/tulu-3-datasets-673b8df14442393f7213f372

Llava o1 - vlm capable of spontaneous, systematic reasoning, similar to GPT-o1, 11B model outperforms gemini-1.5-pro, gpt-4o-mini, and llama-3.2-90B-vision
Xkev/Llama-3.2V-11B-cot

Black Forest Labs Flux.1 tools - four new state of the art model checkpoints & 2 adapters for fill, depth, canny & redux, open weights
reach-vb/black-forest-labs-flux1-6743847bde9997dd26609817

Jina AI Jina CLIP v2 - general purpose multilingual and multimodal (text & image) embedding model, 900M params, 512 x 512 resolution, matroyoshka representations (1024 to 64)
jinaai/jina-clip-v2

Apple AIM v2 & CoreML MobileCLIP - large scale vision encoders outperform CLIP and SigLIP. CoreML optimised MobileCLIP models
apple/aimv2-6720fe1558d94c7805f7688c
apple/coreml-mobileclip

A lot more got released like, OpenScholar (https://huggingface.co/collections/OpenScholar/openscholar-v1-67376a89f6a80f448da411a6), smoltalk ( HuggingFaceTB/smoltalk), Hymba ( nvidia/hymba-673c35516c12c4b98b5e845f), Open ASR Leaderboard ( hf-audio/open_asr_leaderboard) and much more..

Can't wait for the next week! ๐Ÿค—
dvilasueroย 
posted an update 3 months ago
Xenovaย 
posted an update 3 months ago
view post
Post
5723
Have you tried out ๐Ÿค— Transformers.js v3? Here are the new features:
โšก WebGPU support (up to 100x faster than WASM)
๐Ÿ”ข New quantization formats (dtypes)
๐Ÿ› 120 supported architectures in total
๐Ÿ“‚ 25 new example projects and templates
๐Ÿค– Over 1200 pre-converted models
๐ŸŒ Node.js (ESM + CJS), Deno, and Bun compatibility
๐Ÿก A new home on GitHub and NPM

Get started with npm i @huggingface/transformers.

Learn more in our blog post: https://huggingface.co/blog/transformersjs-v3
  • 3 replies
ยท
reach-vbย 
posted an update 3 months ago
view post
Post
4441
What a brilliant week for Open Source AI!

Qwen 2.5 Coder by Alibaba - 0.5B / 1.5B / 3B / 7B / 14B/ 32B (Base + Instruct) Code generation LLMs, with 32B tackling giants like Gemnini 1.5 Pro, Claude Sonnet
Qwen/qwen25-coder-66eaa22e6f99801bf65b0c2f

LLM2CLIP from Microsoft - Leverage LLMs to train ultra-powerful CLIP models! Boosts performance over the previous SOTA by ~17%
microsoft/llm2clip-672323a266173cfa40b32d4c

Athene v2 Chat & Agent by NexusFlow - SoTA general LLM fine-tuned from Qwen 2.5 72B excels at Chat + Function Calling/ JSON/ Agents
Nexusflow/athene-v2-6735b85e505981a794fb02cc

Orca Agent Instruct by Microsoft - 1 million instruct pairs covering text editing, creative writing, coding, reading comprehension, etc - permissively licensed
microsoft/orca-agentinstruct-1M-v1

Ultravox by FixieAI - 70B/ 8B model approaching GPT4o level, pick any LLM, train an adapter with Whisper as Audio Encoder
reach-vb/ultravox-audio-language-model-release-67373b602af0a52b2a88ae71

JanusFlow 1.3 by DeepSeek - Next iteration of their Unified MultiModal LLM Janus with RectifiedFlow
deepseek-ai/JanusFlow-1.3B

Common Corpus by Pleais - 2,003,039,184,047 multilingual, commercially permissive and high quality tokens!
PleIAs/common_corpus

I'm sure I missed a lot, can't wait for the next week!

Put down in comments what I missed! ๐Ÿค—
reach-vbย 
posted an update 3 months ago
view post
Post
1701
Smol TTS models are here! OuteTTS-0.1-350M - Zero shot voice cloning, built on LLaMa architecture, CC-BY license! ๐Ÿ”ฅ

> Pure language modeling approach to TTS
> Zero-shot voice cloning
> LLaMa architecture w/ Audio tokens (WavTokenizer)
> BONUS: Works on-device w/ llama.cpp โšก

Three-step approach to TTS:

> Audio tokenization using WavTokenizer (75 tok per second)
> CTC forced alignment for word-to-audio token mapping
> Structured prompt creation w/ transcription, duration, audio tokens

The model is extremely impressive for 350M parameters! Kudos to the
OuteAI team on such a brilliant feat - I'd love to see this be applied on larger data and smarter backbones like SmolLM ๐Ÿค—

Check out the models here: OuteAI/outetts-6728aa71a53a076e4ba4817c
dvilasueroย 
posted an update 3 months ago
view post
Post
690
Build datasets for AI on the Hugging Face Hubโ€”10x easier than ever!

Today, I'm excited to share our biggest feature since we joined Hugging Face.

Hereโ€™s how it works:

1. Pick a datasetโ€”upload your own or choose from 240K open datasets.
2. Paste the Hub dataset ID into Argilla and set up your labeling interface.
3. Share the URL with your team or the whole community!

And the best part? Itโ€™s:
- No code โ€“ no Python needed
- Integrated โ€“ all within the Hub
- Scalable โ€“ from solo labeling to 100s of contributors

I am incredibly proud of the team for shipping this after weeks of work and many quick iterations.

Let's make this sentence obsolete: "Everyone wants to do the model work, not the data work."


Read, share, and like the HF blog post:
https://huggingface.co/blog/argilla-ui-hub
reach-vbย 
posted an update 3 months ago
view post
Post
3030
Smol models ftw! AMD released AMD OLMo 1B - beats OpenELM, tiny llama on MT Bench, Alpaca Eval - Apache 2.0 licensed ๐Ÿ”ฅ

> Trained with 1.3 trillion (dolma 1.7) tokens on 16 nodes, each with 4 MI250 GPUs

> Three checkpoints:

- AMD OLMo 1B: Pre-trained model
- AMD OLMo 1B SFT: Supervised fine-tuned on Tulu V2, OpenHermes-2.5, WebInstructSub, and Code-Feedback datasets
- AMD OLMo 1B SFT DPO: Aligned with human preferences using Direct Preference Optimization (DPO) on UltraFeedback dataset

Key Insights:
> Pre-trained with less than half the tokens of OLMo-1B
> Post-training steps include two-phase SFT and DPO alignment
> Data for SFT:
- Phase 1: Tulu V2
- Phase 2: OpenHermes-2.5, WebInstructSub, and Code-Feedback

> Model checkpoints on the Hub & Integrated with Transformers โšก๏ธ

Congratulations & kudos to AMD on a brilliant smol model release! ๐Ÿค—

amd/amd-olmo-6723e7d04a49116d8ec95070