Robotics is at an inflection point. We're witnessing a fundamental shift from single-purpose, fixed-function robots to generalist machines that can adapt, reason, and perform diverse tasks across unpredictable environments. This transformation demands something unprecedented: the ability to run massive generative AI models—large language models (LLMs), vision language
I am thrilled to share the release of the Penpot Docker Extension, a tool designed to streamline the deployment and management of a complete self-hosted Penpot instance directly within Docker Desktop.
Quantization = compressing a model by lowering the precision of numbers, making it smaller, faster, and cheaper to run, often with only a small drop in accuracy.
From Claude Desktop to Cursor: A complete breakdown of which AI chat interfaces support MCP—and which ones are worth your time. Because in 2025, your AI assistant should do more than just talk.
No monitor? No problem. Learn how a simple USB Type-C charging cable and serial console access saved a student demo at our Docker meetup.
What if your AI chatbot could configure itself based on what customers ask, without developers editing config files? That's Dynamic MCP.
Want to add vision capabilities to your applications without sending data to external APIs? Docker Model Runner makes it straightforward to run multimodal AI models locally, giving you complete control over your data while using the familiar OpenAI-compatible API format.
Docker
This guide walks you through connecting models from the Docker AI Model Catalog to MCP servers, enabling your applications to leverage both local inference and external capabilities in a secure, reproducible Docker Compose environment.
Stop wasting hours setting up MCP servers. The Docker MCP Catalog provides 270+ enterprise-grade, containerized Model Context Protocol servers that install in seconds—no dependency hell, no environment conflicts, no cross-platform issues.
Docker. Kubernetes. Agentic AI.
How we built, trained, and deployed a dental X-ray analysis system achieving 99.5% mAP50 accuracy using YOLOv8, Docker containers, and iterative model improvement using NVIDIA Jetson AGX Thor
If you're building robots, you're going to want to hear about this.
In this world of influencers, life is not easy. The pressure to stay visible, relevant, and knowledgeable is constant.
Docker Model Runner uses llama.cpp's KV cache for automatic token caching, eliminating redundant prompt processing in local LLM deployments. Discover how this built-in optimization works.
Designed to reduce vulnerabilities and simplify compliance, DHIs integrate easily into your existing Docker-based workflows as well as Kubernetes deployments with little to no retooling required.
Frustrated by tiny context windows when you know your model can handle so much more? If you're running llama.cpp through Docker Model Runner and hitting that annoying 4096 token wall, there's a simple fix you need to know about. Your model isn't the problem—your configuration is.
Post-GITEX, 27 innovators gathered at Coders HQ for the Agentic AI and Security Meetup. The topic? "Agentic AI and Docker." But it was the questions that revealed the real story—engineers already deploying custom models, architects rethinking infrastructure for multi-LLM systems.
Master multi-agent AI workflows with simple YAML configurations—no programming required
Join Docker at GITEX Global 2025—the world's largest tech event—to discover how Docker Hardened Images are revolutionizing container security. With 95% less attack surface, enterprise SLA-backed patching, and built-in compliance, DHI proves security doesn't have to slow you down.
As AI systems evolve from simple chatbots to autonomous agents with memory, tools, and the ability to collaborate, they're creating security vulnerabilities we've never seen before. This deep dive into cutting-edge arXiv research explains why agentic AI is your biggest security challenge.
Your winning strategy for $15,000 in prizes and career opportunities
Vibe coding is the new rhythm of software: start with a fuzzy idea, throw a prompt at an AI, and—boom—a demo runs. The catch? Creation is instant; correctness isn’t. This post unpacks that paradox.
