Stars
A powerful tool for creating fine-tuning datasets for LLM
A Tool to Visualize Claude Code's LLM Interactions
A curated collection of papers and related projects on using LLMs for privacy.
Repository hosting code for "Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations" (https://arxiv.org/abs/2402.17152).
JellyVR is your personal movie theater for Jellyfin, made in Godot Engine 4.4 with C++.
Protect your privacy when using LLM chatbots by letting someone else proxy your query for you.
📚 This is an adapted version of Jina AI's Reader for local deployment using Docker. Convert any URL to an LLM-friendly input with a simple prefix http://127.0.0.1:3000/https://website-to-scrape.com/
Official PyTorch implementation for "Large Language Diffusion Models"
Easiest and laziest way for building multi-agent LLMs applications.
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.
Large Language Model-enhanced Recommender System Papers
A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthropic Claude models.
Official repo for the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas"
verl: Volcano Engine Reinforcement Learning for LLMs
Minimal reproduction of DeepSeek R1-Zero
Robust recipes to align language models with human and AI preferences
Everything about the SmolLM and SmolVLM family of models
🌐 Make websites accessible for AI agents. Automate tasks online with ease.
Performant framework for training, analyzing and visualizing Sparse Autoencoders (SAEs).
[ICML 2024] Binoculars: Zero-Shot Detection of LLM-Generated Text
Autonomous coding agent right in your IDE, capable of creating/editing files, executing commands, using the browser, and more with your permission every step of the way.
Model interpretability and understanding for PyTorch
This repository provides a clear, educational implementation of Byte Pair Encoding (BPE) tokenization in plain Python. The focus is on algorithmic understanding, not raw performance.