Stars
[NeurIPS 2025] L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models
aideate / vllm
Forked from vllm-project/vllmA high-throughput and memory-efficient inference and serving engine for LLMs
Minimal reproduction of DeepSeek R1-Zero
Simulation platform for general-purpose robotics & embodied AI learning.
[CVPR 2025] EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation
Worker to orchestrate and manage running an arbitrary number of LLM-generated builds concurrently using containerized Minecraft Servers.
Text-to-Music Generation with Rectified Flow Transformers
✨✨Latest Advances on Multimodal Large Language Models
Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
Perplexity Inspired Answer Engine
JuiceFS is a distributed POSIX file system built on top of Redis and S3.
Zero-Shot Speech Editing and Text-to-Speech in the Wild
SGLang is a high-performance serving framework for large language models and multimodal models.
📝 Reference-Free automatic summarization evaluation with potential hallucination detection
OCR, layout analysis, reading order, table recognition in 90+ languages
Aligning pretrained language models with instruction data generated by themselves.
Codebase for Merging Language Models (ICML 2024)
[ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding
Machine Learning Engineering Open Book
StyleTTS 2: Towards Human-Level Text-to-Speech through Style Diffusion and Adversarial Training with Large Speech Language Models
The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.
A high-performance inference system for large language models, designed for production environments.
Supercharge Your LLM Application Evaluations 🚀
Reference implementation for DPO (Direct Preference Optimization)