Stars
A lightweight post-training framework for LLMs and VLMs. 51 algorithms, 38 verified models. Scales with DeepSpeed, vLLM, and Ray.
An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex β developed and maintained with no human intervention.
AI-powered terminal manager β control your machines from Telegram. One bot per machine.
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
Teams-first Multi-agent orchestration for Claude Code
Post-training framework for large models, from new objectives to new rollout systems.
Use your telegram to control terminal on remote Linux/MacOS
Autonomous AI development loop for Claude Code with intelligent exit detection
Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.
A configuration framework that enhances Claude Code with specialized commands, cognitive personas, and development methodologies.
ππ€ Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN
A Datacenter Scale Distributed Inference Serving Framework
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
Minimalistic 4D-parallelism distributed training framework for education purpose
Official repo for paper "Structured 3D Latents for Scalable and Versatile 3D Generation" (CVPR'25 Spotlight).
Implement a ChatGPT-like LLM in PyTorch from scratch, step by step
[ICLR24] Official PyTorch Implementation of Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Fast and memory-efficient exact attention
Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
π§βπ« 60+ Implementations/tutorials of deep learning papers with side-by-side notes π; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), ga...
A LLM based research assistant that allows you to have a conversation with a research paper