Llama-3.3-70B-Instruct

Inspired by transformers.js I decided to create a simple and nice UI for whisper + Llama 3.2 1B, fully local in your browser! Check it out!

OpenAI Sora allegedly leaked to Hugging Face

AI Lab PleIAs Releases Fully Open Dataset, as AMD, Ai2 Release Open AI Models

A CC-By Open-Source TTS Model with Voice Cloning

HuggingFace - Tencent launches Hunyuan Large which outperforms Llama 3.1 405B

OpenFLUX.1

nvidia/Llama-3.1-Nemotron-70B-Instruct-HF

Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval (March 22, 2024)

Improving Parquet Dedupe on Hugging Face Hub

Nvidia releases NVLM 1.0 72B open weight model

Llama can now see and run on your device – welcome Llama 3.2

Llama 3.2

Llama-3.2-3B-Instruct-uncensored

Llama 3.1 70B models compressed by 6.4x using state-of-the-art algorithm, now released

Fine-Tuning LLMs to 1.58bit

General OCR Theory: Towards OCR-2.0 via a Unified End-to-End Model

Fake Insects: a game where you have to identify AI-generated insects

HuggingFace to Replace Git LFS with Xet

whisperfile

Perspectives for first principles prompt engineering

FalconMamba 7B: The first attention-free and general-purpose pure Mamba model

Protection against Prompt Injection

: I built a WASM powered Text-to-Speech library that runs in your browser with almost human-like audio quality. DEMO is now available!

Open-LLM performances are plateauing

NuExtract: A LLM for Structured Extraction

CryptGPT: A Simple Approach to Privacy-Preserving LLMs Using Vigenere Cipher

Fine-tune and deploy open LLMs as containers using AIKit - Part 1

Space secrets leak disclosure

FineWeb: Decanting the web for the finest text data at scale

More →