• Fact-checking AI answers: how do you check if they areDiscover how to reliably check the output of language models. Learn practical steps to detect hallucinations and verify facts.
• Using AI Safely at Home: Practical Tips for FamiliesDiscover how to handle artificial intelligence smartly and safely as a family. Practical tips on privacy, agreements with children, and reliable answers.
• When it does not work: out-of-memory, slow tokens and a GPU that stays out of itTroubleshooting article for local LLM setups: out-of-memory, slow tokens and a GPU that stays out of it, with a diagnostic step-by-step plan and solutions per platform.
• Setting up AnythingLLM as a desktop workspaceComplete guide to AnythingLLM as a desktop workplace. Connect local models, manage vector storage, and search documents 100% privately.
• GDPR-compliant with AI: privacy checklist for DutchDiscover how to deploy AI safely and GDPR-compliantly in your organization with this practical privacy checklist.
• Backup strategy for local LLMs & dataA practical 3-2-1 backup strategy for locally run language models, vector databases, prompts, and configurations. Prevent data loss.
• Improving Your AI's Performance in Dutch: A Practical GuideDiscover how to improve the performance of Large Language Models in Dutch. Learn about effective system prompts, tone, register, and avoiding Anglicisms.
• Setting context length for local LLMs | Guide llmnet.nlDiscover how to set context length optimally when running LLMs locally, avoid memory problems, and balance GPU and RAM memory.
• Outsourcing AI tasks smartly through a proxy and routingLearn how, as an indie developer, to delegate programming tasks to specialized models via an OpenAI proxy, including logging and per-task cost control.
• Searching Your Own Documents with Local AI (Private RAG)Discover how to search your own documents and notes privately and locally with AI. Learn all about RAG, tools, hardware, and privacy without the cloud.
• Configuring ExLlamaV2 for maximum EXL2 speedOptimize ExLlamaV2 and EXL2 quantization for maximum token speed on Nvidia GPUs with FlashAttention-2, cache tuning, and speculative sampling.
• Automatically anonymizing personal data with AIAutomatically anonymizing and pseudonymizing personal data with local AI, Microsoft Presidio, NER, and LLMs in compliance with the GDPR.
• What hardware do you need to run LLMs locally?Discover exactly what hardware you need for local LLMs. Learn about RAM, VRAM, GPU versus CPU, and quantization. Practical guide for local AI builders.
• Installing and configuring Jan AI on WindowsGuide to Jan AI on Windows: install the desktop app, configure Vulkan or CUDA GPU offloading, and manage local open-source models.
• Installing Kobold.cpp for local GGUF inferenceA guide to Kobold.cpp: install the standalone GGUF engine, configure GPU offloading and run local language models with context optimization.
• KV cache quantization in llama.cpp and vLLMSave VRAM and extend your context length by quantizing the KV cache to FP8 or INT4 in llama.cpp and vLLM, with concrete runtime parameters and instructions.
• Quantization explained: running large models on smallDiscover what LLM quantization is, how GGUF works, and how to find the ideal balance between model quality and memory usage on modest hardware.
• Compiling llama.cpp with CUDA: maximum performanceLearn step by step how to compile llama.cpp with Nvidia CUDA support for maximum token throughput and minimal latency on a Linux system.
• Configuring GPU Passthrough in Proxmox for LLMsLearn how to configure GPU passthrough in Proxmox VE for local LLMs. A complete step-by-step plan for IOMMU, VFIO, and PCIe assignment.
• Running a local language model (LLM) in DockerLearn why and how to run a local AI language model in Docker. Includes explanations of GPU acceleration, volumes, networks, and performance differences per…
• Running an LLM locally on your own machine | LLM GuideLearn how to easily run a Large Language Model (LLM) locally. Discover the hardware requirements, privacy benefits, and tools like Ollama and LM Studio.
• Running LLMs on Older Hardware: Optimization GuideDiscover how to get usable performance from a local language model on older hardware. Practical guide to RAM management, quantization, and settings.
• Running an LLM on a Synology NASCan you run a Large Language Model locally on a Synology NAS? Discover the possibilities, RAM requirements, and limitations of CPU inference via Docker.
• Deploying LM Studio Headless as a Server | llmnet.nl GuideLearn how to use LM Studio without a graphical user interface as an OpenAI-compatible API server on macOS and Linux. Complete with background services and CLI management.
• Choosing a Local Model: Which Variant Fits Your Goal?Discover how to choose the right local AI model for your goal, such as coding or chatting. Learn all about model size, speed, and Dutch support.
• Local or via an API? The three-year calculationLocal or via an API? A three-year cost calculation: depreciation, electricity, and management versus price per million tokens. Verified on 2026-08-07.
• Running RAG locally: from embedding to answerBuild a complete local RAG pipeline without the cloud. From text parsing and embedding models to vector storage, reranking, and contextual synthesis.
• Local Image Generation: Create AI Images on Your OwnA complete guide to generating images locally with AI. Learn all about hardware requirements, models, software tools, and prompt settings.
• Setting up Continue.dev with local models in VS CodeLearn how to set up Continue.dev in VS Code with local open-weight language models via Ollama for maximum privacy and free code generation.
• Building local function calling with Ollama and PythonBuild reliable local function calling with Ollama and Python. Learn schema validation, Pydantic models, error handling and tool calling pipelines.
• Measuring the energy consumption of your local LLM setupLearn step by step how to measure the exact energy consumption of your local LLM setup at hardware level, through the CLI and with smart plugs.
• Local Fine-Tuning with LoRA: Practical Guide | LLMnetA practical guide to fine-tuning LLMs locally with LoRA. Learn all about data preparation, hyperparameter settings, and evaluating adapters.
• Connecting a Local LLM to VS Code as a Code AssistantLearn how to connect a local AI model to VS Code. Discover the best extensions, settings for Apple Silicon, and tips for privacy-friendly coding.
• Local LLM Servers Secure on Your Network | GuideRead how to configure local LLM servers securely within your network. Prevent unauthorized access, connect a reverse proxy, and set up network segmentation.
• Running Local LLMs on Linux: The Ultimate Guide to OllamaLearn how to run a local LLM on Linux. Step-by-step guide for Ollama, llama.cpp, GPU drivers (NVIDIA/AMD), and headless server configurations.
• Running Local LLMs on Windows with LM Studio and OllamaLearn step-by-step how to install local LLMs like Llama 3 and Mistral on Windows using LM Studio or Ollama. Includes tips for CPU and GPU optimization.
• Local LLMs on Apple Silicon: unified memory, MLX, and Ollama comparedHow unified memory changes the VRAM calculus for local LLMs on Apple Silicon: Ollama, MLX, and LM Studio compared for setup and model selection.
• Local LLMs on Windows/WSL: Installation and DifferencesDiscover whether local LLMs run better on native Windows or through WSL2. Compare performance, installation steps, and hardware integration.
• Securely Accessing a Local LLM Remotely via Tailscale | LLMNet GuideLocalize your AI models at home or at the office and access them securely from anywhere in the world via Tailscale VPN. Step-by-step guide for Ollama and LM Studio.
• Local models in Docker: containers without the hassleRun local AI models in isolation with Docker and GPU passthrough via the NVIDIA Container Toolkit. A complete guide with compose files and optimizations.
• Connecting a local model to your own documents (RAG) on yourLearn how to connect a language model (LLM) to your own documents locally on your Mac using RAG. Secure, with tools like Ollama and vector databases, for…
• Adding local re-ranking to your RAG search pipelineImprove the precision of your local RAG pipeline with cross-encoders and local re-ranking through FlashRank and ONNX without sending data to the cloud.
• Setting up a local vector database: Qdrant and ChromaPractical guide to setting up and configuring Chroma and Qdrant locally. Learn about databases versus files, docker setups, and metadata.
• Running local vision models for image analysisComplete guide to running vision models locally with Ollama and vLLM for OCR, diagram analysis, and privacy-friendly image processing.
• Tracking model updates locally: version control for LLMsLearn how to manage local model versions, GGUF hashes and Ollama tags, prevent update regression and roll back automatically on your own hardware.
• Downloading and Managing Local Models: The Complete GuideLearn how to safely download local LLM models, choose GGUF and quantization formats, and efficiently manage your storage space without chaos.
• Comparing models fairly on your own hardwareA step-by-step plan for testing LLMs objectively on your own hardware: fixing variables, measuring speed and quality, and avoiding faulty comparisons.
• Running Dutch and European open models locallyRunning Dutch and European open models locally: what exists, how to test whether a model really speaks good Dutch. Checked on 2026-08-07.
• Nginx reverse proxy with authentication for your LLMSecure your local LLM server with Nginx, Bearer API keys, TLS encryption, streaming support, and rate limiting against overload.
• Installing Ollama and running your first local model onDiscover step-by-step how to install Ollama and run local language models on macOS. Includes hardware requirements, terminal commands, and useful tips.
• Editing an Ollama Modelfile: system prompt and parametersLearn how to set the system prompt and parameters with an Ollama Modelfile. An extensive guide with examples, parameter explanations and management.
• Setting up Open WebUI: a pleasant chat interface for yourDiscover how to set up Open WebUI as a user-friendly chat interface for your local LLMs via Ollama. Practical guide for installation and configuration.
• Power Limits and Fan Curves for GPU ServersPractical guide to power limits, persistence mode, and fan curves on Linux GPU servers for energy-efficient, quiet, and thermally stable AI inference.
• Using AI in a Privacy-Friendly Way in Daily WorkPractical guide for safe AI use at work. Discover which data you should keep local, essential privacy settings, and free privacy-conscious tools.
• Quantization in Practice: Compressing Models YourselfLearn step by step how to convert open-source language models to GGUF and quantize them yourself using llama.cpp for optimal local performance.
• Configuring speculative decoding for faster LLM tokensPractical guide to configuring speculative decoding in llama.cpp, vLLM, and Ollama with draft models for maximum token throughput.
• Converting Speech to Text LocallyLearn how to convert speech to text locally and privacy-friendly on your own Mac. Discover the best models, hardware requirements, and diarization techniques.
• How Much Does the Power Consumption of Local AI Cost? (GPU/CPU CalculationCalculate exactly what the power consumption of local AI models costs. Practical formulas for GPU/CPU inference, idle consumption, and a cloud comparison.
• Enforcing Structured JSON Outputs in Local LLMsEnforce valid JSON in local LLMs: JSON schemas in Ollama, LM Studio, and vLLM, GBNF grammars in llama.cpp, plus schema design and error handling.
• Installing Tabby as a local code assistant with DockerLearn step by step how to install Tabby as a self-hosted AI code assistant with Docker and GPU support for VS Code and JetBrains.
• Installing Text Generation WebUI on LinuxInstall Text Generation WebUI (oobabooga) on a Linux server. Practical guide covering CUDA, loaders, CLI flags, and network configuration.
• Combining Two GPUs for Local Language ModelsPractical guide to dual-GPU setups: tensor and pipeline parallelism, PCIe bandwidth, VRAM pooling, and configuration in llama.cpp and vLLM.
• Configuring vLLM for high throughput on LinuxRead how to set up vLLM on Linux for optimal throughput with multiple concurrent requests. A guide to memory management, batching and security.
• VRAM calculator: how much memory does a model need?Calculate exactly how much VRAM a local LLM requires. Formulas for model parameters, quantization, context length, KV cache, and CUDA overhead.
• Choosing an embedding model for Dutch documentsFind out which embedding model performs best for Dutch-language documents in a local RAG setup. With benchmarks, memory requirements, and practical tips.