We built an AI agent with Hugging Face's smolagents and a Model Context Protocol server to make Jaqpot models accessible in AI workflows and assistants like Claude Desktop.

In the era of the European Green Deal, the push toward safer and more sustainable chemicals and materials is stronger than ever. Jaqpot, our cloud-based modelling platform, is uniquely positioned to meet this challenge by integrating state-of-the-art predictive models with automation capabilities.

Download your Jaqpot models and run predictions locally without internet connectivity. Perfect for production environments, data privacy requirements, and air-gapped systems.

This post explains how you can train and upload a standard PyTorch regression model

Want to deploy your LLM with real-time streaming responses? Here's a complete guide covering backend, frontend, and deployment.

Learn how to create a graph-based classification model using molecular structures to predict AMES mutagenicity. This guide walks through the process from data preparation to deploying the model on Jaqpot's platform.

Want to deploy your LLM with real-time streaming responses? Here's a complete guide covering backend, frontend, and deployment.

Learn how to create datasets with jaqpotpy. This guide covers simple numeric data, SMILES-based features, and task setup for regression or classification.

Ready to deploy your first machine learning model? This guide walks you through creating and deploying a simple classification model using Jaqpot. We'll show you how to prepare your data, train a model, and deploy it with just a few lines of code.

Welcome to our very first blog post! We are excited to finally share our story, our mission, and the journey that led to the creation of Jaqpot. Whether you're a machine learning enthusiast, a researcher, or just curious about who we are, this post will give you a peek behind the scenes.