Async RL Training at Scale

[VLDB' 25] Synthesizing High-quality Text-to-SQL Data at Scale. SynSQL-2.5M is the first million-scale cross-domain text-to-SQL dataset.

minimal-cost for training 0.5B R1-Zero

A dataset of poker hand histories

A framework for few-shot evaluation of language models.

Official Implementation for "ATP: All-in-one tuning and structural pruning for domain-specific llms"

LLaMA-Omni is a low-latency and high-quality end-to-end speech interaction model built upon Llama-3.1-8B-Instruct, aiming to achieve speech capabilities at the GPT-4o level.

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

AdaCode is licensed under CC BY-NC-SA 4.0 https://creativecommons.org/licenses/by-nc-sa/4.0/

[CVPR2023] The implementation for "DiffTalk: Crafting Diffusion Models for Generalized Audio-Driven Portraits Animation"

Large-scale text-video dataset. 10 million captioned short videos.

Stable Diffusion web UI

AnimateDiff I2V version.

A retrain of AnimateDiff to be conditional on an init image

A PyTorch implementation of MAGE: MAsked Generative Encoder to Unify Representation Learning and Image Synthesis

PyTorch implementation of RCG https://arxiv.org/abs/2312.03701

High-efficiency floating-point neural network inference operators for mobile, server, and Web