ai projects

7 European open-source projects in the ai category.

Aleph Cloud

PythonMIT

Aleph Cloud is a decentralized SuperCloud that provides compute, storage, and frontend hosting on independent infrastructure. It enables projects to run high-availability backends and verifiable, censorship-resistant frontends, acting as a resilient alternative or backup to centralized clouds like AWS or Cloudflare. Aleph Cloud is used by Web3 and also web2 companies, DeFi, and AI projects to improve resilience, decentralization, and trust, with pay-as-you-go usage and blockchain-native integrations.

Anemoi Core

PythonApache-2.0

Core packages for ECMWF Anemoi machine learning weather forecasting framework.

ecmwf

Cheshire Cat AI

PythonGPL-3.0

Python framework to build AI agents and publish them on the web. Multi-user, multi-agent, extensible via plugins, supports MCP and A2A protocols.

ColiZig

OtherMIT

A from-scratch Zig inference engine for large MoE models - stream experts from SSD to RAM to VRAM and keep the resident working set small. ColiZig explores a different way to run Mixture-of-Experts models: instead of loading the entire model into memory, it treats expert weights as a tiered working set and streams them between SSD, RAM and opt. VRAM

datapizza-ai

PythonMIT

datapizza-ai is a Python GenAI framework for building agents from development to production with minimal abstraction. It supports multi-provider LLMs (OpenAI, Anthropic, Gemini, Mistral), built-in OpenTelemetry tracing, modular RAG pipelines, and multi-agent orchestration. Built by the Italian Data & AI community Datapizza.

Robotoff

PythonAGPL-3.0

Machine learning prediction service for Open Food Facts product categorization and quality checks.

open-food-facts

ScrapeGraphAI

PythonMIT

ScrapeGraphAI is a Python library for AI-powered web scraping that uses LLMs to extract structured data from websites, PDFs, and raw HTML via natural language queries. Built by an Italian team in Padua, it supports OpenAI, Gemini, Ollama, and Anthropic, eliminating the need for fragile CSS or XPath selectors.