arhsis
Follow
Lists (1)
Stars
Python tool for converting files and office documents to Markdown.
A high-throughput and memory-efficient inference and serving engine for LLMs
AI agents running research on single-GPU nanochat training automatically
Unsloth Studio is a web UI for training and running open models like Qwen3.5, Gemma 4, DeepSeek, gpt-oss locally.
The simplest, fastest repository for training/finetuning medium-sized GPTs.
Rich is a Python library for rich text and beautiful formatting in the terminal.
🚀🚀 「大模型」2小时完全从0训练64M的小参数GPT！🌏 Train a 64M-parameter GPT from scratch in just 2h!
"🐈 nanobot: The Ultra-Lightweight Personal AI Agent"
A book-in-progress about the Linux kernel and its insides.
SGLang is a high-performance serving framework for large language models and multimodal models.
Machine Learning Engineering Open Book
Code samples for my book "Neural Networks and Deep Learning"
The official GitHub page for the survey paper "A Survey of Large Language Models".
Self-hosted bookmark manager that is designed be to be minimal, fast, and easy to set up using Docker.
My learning notes for ML SYS.
PDF craft can convert PDF files into various other formats. This project will focus on processing PDF files of scanned books.
Minimal Claude Code alternative. Single Python file, zero dependencies, ~250 lines.
Pyzotero: a Python client for the Zotero API
LLM notes, including model inference, transformer model structure, and llm framework code analysis notes.
Serverless LLM Serving for Everyone.
Puzzles for learning Triton, play it with minimal environment configuration!
A tiny yet powerful LLM inference system tailored for researching purpose. vLLM-equivalent performance with only 2k lines of code (2% of vLLM).
A Telegram bot to recommend arXiv papers