Локальные LLM
litellm
The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, an…
★ 60,1 тыс. звёздCustom
Смотреть руководство →gemma — проект с открытым исходным кодом из категории «Локальные LLM» от google-deepmind (google-deepmind/gemma на GitHub). У него 5 774 звезды на GitHub и 1 032 форка, он распространяется по лицензии Apache-2.0 и последний раз обновлялся на GitHub 6 октября 2026.
Google DeepMind's Gemma open-weight LLM library — install, load and chat with Gemma models in Python.
Decide which Gemma model size fits your hardware and whether you need text-only or multimodal input.
Install the gemma package from PyPI into your Python environment.
Get the model weights from the official Gemma download channels and note their license terms.
Load the model with the example code and start a multi-turn chat.
The reference Python library for Gemma models, maintained by Google DeepMind.
The bundled example runs a multi-turn, multimodal conversation with Gemma 4-class models.
pip install gemma gets the library; weights are fetched separately so code and models update independently.
Links to the full documentation site and the technical reports describing model capabilities and metrics.
Points to other Gemma implementations and tooling across frameworks if you do not run JAX.
Gemma models are open-weight releases from Google, usable locally under the Gemma terms.
Clone the Gemma library repository.
git clone https://github.com/google-deepmind/gemma.git
cd gemmaInstall the gemma package from PyPI.
pip install gemmaDownload the weights as the README directs, then run the minimal multi-turn conversation example.
python -m gemma.examples # see README for the exact commandGoogle DeepMind maintains the gemma repository and the Gemma model family.
The library is Apache-2.0 licensed; the model weights are open weights distributed under Google's Gemma terms, which you accept when downloading them.
No — you run the weights locally. You do need to download the weights through the official channels, which may require accepting the Gemma terms.
This is DeepMind's own JAX-based reference implementation; the Transformers route is often easier if your stack is PyTorch, but this repo is the canonical source for the library and examples.
pip install gemma, download the weights per the README, and run the conversation example.
The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, an…
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
A no-install GUI for Ollama and other local LLM providers.
Open bilingual (Chinese-English) 6B chat LLM with 32K-context variant and low-memory INT4 inference.
Access large language models from the command-line
Typescript/React Library for AI Chat 💬🚀
Другие проекты из категории «Локальные LLM» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если gemma вам не подходит.
The simplest way to run open language models on your own machine.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
LLM inference in C/C++
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.
Соседние проекты из категории «Локальные LLM» в каталоге, ближайшие по числу звёзд к gemma — продолжайте изучать полку по обе стороны от него.
Discover and run local AI apps, serving optimized LLMs from your own GPU and NPU.
Open-source AI sales OS — self-hosted CRM with native AI agents + WhatsApp (WAHA). Open alternative to Kommo, Octadesk & Intercom for any business that sells…
⚡ Pure-Rust WebGPU inference engine — OpenAI-API compatible, GGUF native, runs on any GPU. No Python. No llama.cpp. Single binary.
GGUF Quantization support for native ComfyUI models
Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots
RamaLama is an open-source developer tool that simplifies the local serving of AI models from any source and facilitates their use for inference in productio…
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.