Stars
Causal depthwise conv1d in CUDA, with a PyTorch interface
MiniMax-H3 全模态视频生成模型部署方案 — RTX Pro 6000 Blackwell 单卡实践
Palm on-device acceleration capabilities.
Ongoing research training transformer models at scale
Stable and Efficient Reinforcement Learning for Trillion-Parameter LLMs
SGLang is a high-performance serving framework for large language models and multimodal models.
Official repository of PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective
🔥 LeetCode for PyTorch — practice implementing softmax, attention, GPT-2 and more from scratch with instant auto-grading. Jupyter-based, self-hosted or try online.
Machine Learning Engineering Open Book
😼 优雅地使用基于 clash/mihomo 的代理环境
Official implementation of "Controlling Text-to-Image Diffusion by Orthogonal Finetuning".
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
使用Gradio为ModRWKV-VLM + RWKV LLM 组合创建了一个可视化界面Demo 🤗
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
An early research stage expert-parallel load balancer for MoE models based on linear programming.
This repository contains an implementation of SliceFine, a parameter-efficient fine-tuning (PEFT) method proposed in the paper titled "SliceFine The Universal Winning-Slice Hypothesis For Pre-train…
PeRL: Parameter-Efficient Reinforcement Learning
A PyTorch native platform for training generative AI models
This is the official repository for C3-OWD: A Curriculum Cross-modal Contrastive Learning Framework for Open-World Detection
A high-throughput and memory-efficient inference and serving engine for LLMs
GLM-4.6V/4.5V/4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
Text-audio foundation model from Boson AI