Hacker Newsnew | past | comments | ask | show | jobs | submit | fromlogin
Meta's Omnilingual ASR: Open-Source Speech Recognition for 1600 Languages (github.com/facebookresearch)
2 points by petralithic 7 days ago | past | 1 comment
Single-Step Diffusion Decoder for Efficient Image Tokenization (github.com/facebookresearch)
1 point by lnyan 39 days ago | past
SSDD: Single-Step Diffusion Decoder for Efficient Image Tokenization (github.com/facebookresearch)
3 points by montyanderson 42 days ago | past
MapAnything: Universal Feed-Forward Metric 3D Reconstruction (github.com/facebookresearch)
2 points by simonpure 60 days ago | past
Facebook Research releases MapAnything, 3D reconstruction from images (github.com/facebookresearch)
2 points by mustaphah 61 days ago | past
DINOv3 (github.com/facebookresearch)
179 points by reqo 3 months ago | past | 31 comments
Code for BLT research paper: Meta find way to bypass tokenization (github.com/facebookresearch)
1 point by kiyanwang 3 months ago | past
V-JEPA 2: Self-Supervised Video Models Enable Understanding,Prediction,Planning (github.com/facebookresearch)
6 points by johlo 4 months ago | past
LLM Speedrunner: Eval for frontier models to reproduce scientific findings (github.com/facebookresearch)
2 points by zerojames 4 months ago | past
Fastgen – SOTA LLM inference in 3k lines of Python (github.com/facebookresearch)
3 points by mpu 6 months ago | past | 1 comment
LLMs can see and hear without any training (github.com/facebookresearch)
210 points by T-A 6 months ago | past | 66 comments
VGGT: Visual Geometry Grounded Transformer (github.com/facebookresearch)
190 points by xnx 7 months ago | past | 42 comments
Facebook Uncommon Objects in 3D Dataset (github.com/facebookresearch)
1 point by taikon 10 months ago | past
Large Concept Models: Language modeling in a sentence representation space (github.com/facebookresearch)
167 points by batata_frita 10 months ago | past | 58 comments
Video Seal: Open and Efficient Video Watermarking (github.com/facebookresearch)
1 point by cloudking 11 months ago | past
Solve Global Lyapunov functions with symbolic transformers (github.com/facebookresearch)
2 points by besart_hoxhaj 11 months ago | past
Watermark Anything (github.com/facebookresearch)
314 points by zerojames on Nov 12, 2024 | past | 108 comments
LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding (github.com/facebookresearch)
1 point by zerojames on Oct 18, 2024 | past
Meta Lingua: Lean, efficient, and easy-to-hack codebase to research LLMs (github.com/facebookresearch)
1 point by zerojames on Oct 18, 2024 | past
Foundation for Human Vision Models (github.com/facebookresearch)
110 points by yoknapathawa on Aug 24, 2024 | past | 21 comments
Nougat – Pdf to Markdown (github.com/facebookresearch)
2 points by lijunhao on Aug 16, 2024 | past | 1 comment
VFusion3D: Learning Scalable 3D Generative Models from Video Diffusion Models (github.com/facebookresearch)
56 points by MrTrvp on Aug 9, 2024 | past | 5 comments
Unibench: Vision-Language Model Evaluation (github.com/facebookresearch)
1 point by zerojames on Aug 8, 2024 | past
SAM 2: Segment Anything in Images and Videos (github.com/facebookresearch)
824 points by xenova on July 29, 2024 | past | 147 comments
MobileLLM: Optimizing Sub-Billion Parameter Language Models for On-Device Use (github.com/facebookresearch)
260 points by tosh on July 9, 2024 | past | 55 comments
Data2vec 2.0 (github.com/facebookresearch)
1 point by Bluestein on July 1, 2024 | past
FAIR's Chameleon Early Fusion Code (github.com/facebookresearch)
1 point by ashvardanian on June 21, 2024 | past | 1 comment
Repository for Meta Chameleon a mixed-modal early-fusion foundation model (github.com/facebookresearch)
2 points by yeldarb on June 18, 2024 | past
Densely Captioned Images (DCI) Dataset (github.com/facebookresearch)
1 point by zerojames on June 5, 2024 | past
STMG: A Machine Learning Microgesture Recognition System (github.com/facebookresearch)
2 points by zerojames on May 15, 2024 | past

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: