LLM reads a paper and produce a working prototype

Train, tune, and infer Bamba model

Code for Quiet-STaR

PDF scientific paper translation with preserved formats - 基于 AI 完整保留排版的 PDF 文档全文双语翻译，支持 Google/DeepL/Ollama/OpenAI 等服务，提供 CLI/GUI/Docker

Learning to Compress Prompts with Gist Tokens - https://arxiv.org/abs/2304.08467

The website for PyTorch

Puzzles for learning Triton, play it with minimal environment configuration!

Modeling, training, eval, and inference code for OLMo

Material for gpu-mode lectures

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

[NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models

Run your own AI cluster at home with everyday devices 📱💻 🖥️⌚

Tiny PyTorch library for maintaining a moving average of a collection of parameters.

Implementation of the proposed minGRU in Pytorch

build ai agents that have the full context, open source, runs locally, developer friendly. 24/7 screen, mic, keyboard recording and control

[EMNLP 2023] Adapting Language Models to Compress Long Contexts

Using FlexAttention to compute attention with different masking patterns

A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.

Source code of Telegram for macos on Swift 5.0

⚡️HivisionIDPhotos: a lightweight and efficient AI ID photos tools. 一个轻量级的AI证件照制作算法。

NVIDIA Cosmos Nemotron is a family of state-of-the-art vision language models (VLMs) for diverse multimodal AI tasks across the edge, data center, and cloud.

An automated pipeline for evaluating LLMs for role-playing.

Extend existing LLMs way beyond the original training length with constant memory usage, without retraining