Facebookresearch

A Python framework for AI-driven character animation using neural networks.

🎬ActionMesh: A fast video to animated mesh model with unprecedented quality. Generate animated mesh seamlessly importable into any 3D software in less than a minute.

AIRS-Bench: an AI Research Science benchmark for quantifying the end-to-end AI research abilities of LLM agents

AssemblyHands Toolkit is a Python package that provides data loader, visualization, and evaluation tools for the AssemblyHands dataset (CVPR 2023).

Watermarking and detection for speech audios

Train universal codec avatars

Comprehensive benchmark for RAG

CoTracker is a model for tracking any point (pixel) on a video.

Training Large Language Model to Reason in a Continuous Latent Space

Large dataset of hand-object contact, hand- and object-pose, and 2.9 M RGB-D grasp images.

State-of-the-Art AI Watermarking

Code release for "Cut and Learn for Unsupervised Object Detection and Instance Segmentation" and "VideoCutLER: Surprisingly Simple Unsupervised Video Instance Segmentation"

Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

Filtering, Distillation, and Hard Negatives for Vision-Language Pre-Training

Dr. Zero Self-Evolving Search Agents without Training Data

Drivable Volumetric Avatars

visualization code for 3D human body annotation by EFT (Exemplar Fine-tuning)

[CVPR 2025] Official PyTorch implementation of "EdgeTAM: On-Device Track Anything Model"

Ego4d dataset repository. Download the dataset, visualize, extract features & example usage of the dataset

Demo utilities for running EgoBlur person and license plate blurring models.

Execution and caching tool for python

PyTorch extensions for high performance and large scale training.

A modular embodied agent architecture and platform for building embodied agents

Fairseq is a sequence modeling toolkit for training custom models in text generation tasks.

A library for efficient similarity search and clustering of dense vectors.

GPU Cluster Monitoring (GCM): Large-Scale AI Research Cluster Monitoring

Geometric Retargeting A Principled, Ultrafast Neural Hand Retargeting Algorithm

Habitat-Lab: a modular high-level library for end-to-end development in Embodied AI.

A flexible, high-performance 3D simulator for Embodied AI research.

A python library that provides common I/O interface across different storage backends.

Kats is a Python toolkit for analyzing time series data, offering tools for detection, forecasting, and feature extraction.

The Automated LLM Speedrunning Benchmark measures how well LLM agents can reproduce previous innovations and discover new ones in language modeling.

Code repo for the paper "LLM-QAT Data-Free Quantization Aware Training for Large Language Models"

Large Concept Models: Language modeling in a sentence representation space

Code for "LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", ACL 2024

Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

Understanding Training Dynamics of Deep ReLU Networks

Matrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data generation.

Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, sparsely activated memory layers complement compute-heavy dense feed-forward layers, providing dedicated capacity to store and retrieve…

Repository for the CVPR 2026 paper MeshFlow Efficient Artistic Mesh Generation via MeshVAE and Flow-based Diffusion Transformer by Weiyu Li, Antoine Toisoul, Tom Monnier, Roman Shapovalov, Rakesh Ranjan, Ping Tan and Andrea Vedaldi.

Mobile vision models and code

[CVPR 2026] Multi-SpatialMLLM: Multi-Frame Spatial Understanding with Multi-Modal Large Language Models

Ocean is the in-house framework for Computer Vision (CV) and Augmented Reality (AR) applications at Meta. It is platform independent and is mainly implemented in C/C++.

Source Code for Paper "OrienterNet Visual Localization in 2D Public Maps with Neural Matching"

PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for evaluation of training and inference platforms.

This repository contains the training code of ParetoQ introduced in our work "ParetoQ Scaling Laws in Extremely Low-bit LLM Quantization"

State-of-the-art Image & Video CLIP, Multimodal Large Language Models, and More!

[CVPR 2026] Pixio: a capable vision encoder dedicated to dense prediction, simply by pixel reconstruction

projectaria_tools is an C++/Python open-source toolkit to interact with Project Aria data.

This repository provides the code for implementing RPG described in our KDD'25 paper "Generating Long Semantic IDs in Parallel for Recommendation".

Repository for research in the field of Responsible NLP at Meta.

Superintelligent Retrieval Agent (SIRA)

SONAR, a new multilingual and multimodal fixed-size sentence embedding space, with a full suite of speech and text encoders and decoders.

Official repository for "Revisiting Weakly Supervised Pre-Training of Visual Perception Models". https://arxiv.org/abs/2201.08371.

Foundational Models for State-of-the-Art Speech and Text Translation

PySlowFast: video understanding codebase from FAIR for reproducing state-of-the-art video models.

Scalable and Performant Data Loading

A general physic-based retargeting framework.

Code repo for the paper "SpinQuant LLM quantization with learned rotations"

Simple yet SoTA Knowledge Graph Embeddings.

Official implementation of the paper "The Stable Signature Rooting Watermarks in Latent Diffusion Models"

Large-Scale Translation Data Mining.

Benchmark and research code for the paper SWEET-RL Training Multi-Turn LLM Agents onCollaborative Reasoning Tasks

A library for differentiable nonlinear optimization

Official implementation of paper "VLM³: Vision Language Models Are Native 3D Learners".

Open and efficient video and image watermarking

An Analysis Toolkit for Natural Language Generation (Translation, Captioning, Summarization, etc.)

Code repository for emg2pose dataset and model benchmarks

FAIR Sequence Modeling Toolkit 2

This repository contains the code to train and evaluate TRIBE v2, a multimodal model for brain response prediction