Локальные LLM
llm
Access large language models from the command-line
★ 12,6 тыс. звёздApache-2.0
Смотреть руководство →llama.cpp — проект с открытым исходным кодом из категории «Локальные LLM» от ggml-org (ggml-org/llama.cpp на GitHub). У него 130 229 звёзд на GitHub и 24 047 форков, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 3 октября 2026.
LLM inference in C/C++
One binary or app gets you a working local model setup.
Pick a size that fits your RAM/VRAM; smaller models run on laptops.
Use the built-in UI, or point OpenAI-compatible clients at it.
Airplane mode works — your prompts never leave the device.
Existing tools plug in by changing one URL.
Ask questions over your PDFs, notes, and codebases.
Pay nothing per token, ever.
Prompts and documents stay on your hardware, always.
RAG, tools, and agents build on the same base.
Smart compression runs big models on modest hardware.
Download the source and move into the project folder.
git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cppEvery project documents its own setup; start with install requirements.
cat README.md | head -60Follow the project's launch command — Docker, pip, npm, or a binary release.
# see README for the exact launch commandAlso on this shelf: api-relay-audit — local security auditor for AI API relays — catches injection and swaps. The Local LLM shelf also holds agents — open-source marketing agents for Claude, ChatGPT, and Codex. The Local LLM shelf also holds ai-text-detector — a cautious, explainable AI-like text risk analysis tool for local workflows and coding.
It is an open-source project at ggml-org/llama.cpp with 24,047 forks. Check the repo's contributors and recent commits to see how active maintenance is.
MIT licensed — use it commercially, no strings attached.
Close on many tasks now; frontier cloud models still lead on hard reasoning.
8B models run on 16GB RAM; bigger models want a real GPU.
A recent 8–32B instruct model is the sweet spot for most machines.
Access large language models from the command-line
Official repo for spec & SDK of MCP Apps protocol - standard for UIs embedded AI chatbots, served by MCP servers
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
⚡ Pure-Rust WebGPU inference engine — OpenAI-API compatible, GGUF native, runs on any GPU. No Python. No llama.cpp. Single binary.
A fast LLM inference server for Apple Silicon, managed from the macOS menu bar
Open bilingual (Chinese-English) 6B chat LLM with 32K-context variant and low-memory INT4 inference.
Другие проекты из категории «Локальные LLM» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если llama.cpp вам не подходит.
The simplest way to run open language models on your own machine.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
A high-throughput and memory-efficient inference and serving engine for LLMs
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
GPT4All — run local LLMs on any device: a free desktop chat client and Python SDK that run open models entirely on your CPU, no GPU or cloud required
Соседние проекты из категории «Локальные LLM» в каталоге, ближайшие по числу звёзд к llama.cpp — продолжайте изучать полку по обе стороны от него.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
A high-throughput and memory-efficient inference and serving engine for LLMs
The simplest way to run open language models on your own machine.
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
Find the best local models for your hardware — and how to run them.
GPT4All — run local LLMs on any device: a free desktop chat client and Python SDK that run open models entirely on your CPU, no GPU or cloud required
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.