#llm Repositories
2,314 repositories total
A high-performance ML model serving framework, offers dynamic batching and CPU/GPU pipelines to fully exploit your compute machine
A production-ready template to kickstart your Generative AI projects with structure and scalability in mind.
JevChat-Windows: A reply assistant docked beside the chat window. It captures a screenshot of the window and uses local, offline OCR to read the other personโs messages โ Jev determines the intent โ 3 suggested replies can be inserted with one click; sending is always manual.
โ๏ธGenAI powered multi-agentic medical diagnostics and healthcare research assistance chatbot. ๐ฅ Designed for healthcare professionals, researchers and patients.
LAYRAโan enterprise-ready, out-of-the-box solutionโunlocks next-generation intelligent systems powered by visual RAG and limitless visual multi-step agent workflow orchestration.
LLM notes, including model inference, transformer model structure, and llm framework code analysis notes.
An LLM ASR voice input method keyboard application for the Android platform based on Kotlin
AI-powered tools to enhance Anki flashcards with explanations, mnemonics, illustrations, and adaptive learning for medical school and beyond
A talking LLM that runs on your own computer without needing the internet.
Code and implementations for the paper "AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning" by Zhiheng Xi et al.
A PyTorch library of curated Transformer models and their composable components
A local, offline (after initial setup), portable OCR software that can process images and PDF files, using DeepSeek-OCR-2 AI (running directly on your machine).
A copy-paste system prompt that has your coding agent (Claude Code, Codex, Cursor) build a self-improving agent system for any computer work. Try the lead follow-up demo at the website link.
HERO = Hashing ยท Edge cases ยท Rubrics ยท Overbuild โ the four shapes coding agents over-defend in. A paste-in contract that stops them. Works with Claude Code, Codex, Antigravity, Cursor, Copilot, Windsurf, Gemini CLI.
Code2Prompt is a powerful command-line tool that simplifies the process of providing context to Large Language Models (LLMs) by generating a comprehensive Markdown file containing the content of your codebase. โญ If you find Code2Prompt useful, consider giving us a star on GitHub! It helps us reach more developers and improve the tool. โญ
๐ฆ๏ธ A fast, secure MCP server that extends its capabilities through WebAssembly plugins.
LLM.swift is a simple and readable library that allows you to interact with large language models locally with ease for macOS, iOS, watchOS, tvOS, and visionOS.
A high-performance framework for training LLMs, VLMs, diffusion, and embodied models on NVIDIA GPUs and Kunlun XPUs. Supports Pi0.5, GR00T-N1.7, FastWAM, Cosmos3, DreamZero, DeepSeek-V4, GLM-5.3, Kimi-K3, and more.
oh-my-zsh for Claude Code โ 16 agents, 35 commands, 32 skills, 21 safety hooks in one install. v4.0 adds an adversarial review loop: a second agent that never sees the first one's reasoning. MIT.
This is an MCP related to the financial field. This project integrates Tushare API and Binance API to provide language models (such as Claude) with comprehensive real-time financial data access capabilities, supporting multi-dimensional financial data analysis including stocks, funds, bonds, macroeconomic indicators, stablecoins, cryptocurrencies, etc. It also includes financial data queries, financial news queries, National Bureau of Statistics data queries, and more.
[EMNLP 2026] MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research ยท Browser-hosted Android Simulator ยท Verifiable Evaluation ยท Scalable Online RL Training
Interactive CV with AI chat integration. Built with React 19, TypeScript, Claude API. Chat with my AI avatar about my experience.
๐ฅMachine Learning/Deep Learning/Python/Large Models/Multimodal/LLM/Deep Learning/Python/Algorithm Interview/NLP Tutorial
OpenSwarm โ Autonomous AI dev team orchestrator powered by Claude Code CLI. Discord control, Linear integration, cognitive memory.
Extract Douyin watermark-free video links, video captions, douyin-mcp-server, mcp, claude skill, supports lobster
llama.cpp on a jailbroken PS5: run any GGUF model natively on the console, starting with Qwen 3.8
Halo is an open-source framework built by White Circle for training large language and multimodal models
xpander.ai is the runtime and control plane to build, run, and ship reliable AI agents fast and anywhere
GLM-ASR-Nano: A robust, open-source speech recognition model with 1.5B parameters
Neum AI is a best-in-class framework to manage the creation and synchronization of vector embeddings at large scale.
Open-source enforcement at runtime that masks sensitive data and blocks destructive commands from AI agents. You change one thing: your agent's port.
Browser automation system that uses AI-driven planning to navigate web pages and perform goals.
A platform for building and operating digital employees for enterprises: codify the experience, workflows, and decision-making standards of skilled employees into AI digital employees that are ready to work at any time, configurable, subject to approval, and observable.
Code and implementations for the ACL 2025 paper "AgentGym: Evolving Large Language Model-based Agents across Diverse Environments" by Zhiheng Xi et al.
[ICLR 2025] Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing. Your efficient and high-quality synthetic data generation pipeline!
End-to-end RAG system design, evaluation, and optimization. GeekTime's RAG bootcamp deconstructs all 10 major components of RAG, with four handsโon projects to master the entire RAG workflow. The adoption of RAG typically focuses on applying RAG to business, rather than building business for RAG. That's why we need to tailor adjustments, optimizations, and customizations for different scenarios and issues. The devil is in the details; we delve deep to investigate.
The artifact layer for any agent: what your AI writes, checked against your files. A free, open-source desktop app for macOS, Windows and Linux. Connect Claude Code or another agent, or use your ChatGPT plan, your own key or a local model. Your files in one Library, sorted and searchable, with your meetings.