Ollama Guides: Local LLM Installation, Configuration, and Integration
19 posts in this series
Use this series when you want to run LLMs locally on your own hardware without relying on cloud APIs. It runs from basic installation to model tuning, API setup, Web UI, and integration with RAG frameworks.
Getting Started with Ollama: Your First Step to Running LLMs Locally
Want to run large language models on your own machine? This guide walks you through installing and configuring Ollama from scratch, covering multi-platform setup, model management, GPU acceleration, and API integration
Complete Guide to Ollama Model Management: Download, Switch, Delete & Version Control
Master Ollama model management with pull, run, list, rm commands. Learn version selection, batch deletion scripts, disk space optimization. Perfect for AI developers and OpenClaw deployers managing local LLM libraries.
Ollama Version Rollback Guide: 3 Critical Steps 90% of Developers Overlook
System unstable after Ollama upgrade? This article provides three complete rollback solutions (binary replacement, package manager, Docker), one-click automation scripts, and a practical multi-version coexistence guide to help you quickly solve version management challenges.
Ollama Modelfile Parameters Explained: A Complete Guide to Creating Custom Models
A detailed guide to Ollama Modelfile's 10 core parameters, including optimization tips for temperature, num_ctx, and more. Includes 4 ready-to-use practical templates to help you create your own custom models.
Running Llama 70B Locally: Comparison and Selection Guide for 5700XT, Mac M4, and CUDA Solutions
Want to run Llama 70B locally? This article compares three solutions: AMD 5700XT, Mac M4, and NVIDIA CUDA, using real test data to help you determine which hardware suits your needs, including VRAM requirements, performance comparisons, and troubleshooting tips.
Ollama Hardware Selection Guide: VRAM, Quantization & GPU Comparison (2026)
A comprehensive Ollama hardware selection table: VRAM requirements for 7B/13B/70B models, Q4/Q8 quantization comparison, and detailed breakdown of NVIDIA CUDA/AMD ROCm/Apple Metal acceleration. Tiered GPU recommendations from RTX 3060 to 5090 help you match the right graphics card to your model needs.
Ollama GPU Acceleration: Complete Guide for CUDA, ROCm & Metal
Complete Ollama GPU acceleration guide covering NVIDIA CUDA, AMD ROCm, and Apple Metal platforms. Includes verification steps, multi-GPU setup, and troubleshooting for 10-20x faster local LLM inference.
Ollama GPU Acceleration Configuration: CUDA, ROCm, and Metal Platform Guide
Comprehensive guide to Ollama GPU acceleration configuration covering NVIDIA CUDA, AMD ROCm, and Apple Metal platforms. Includes hardware requirements, driver installation, verification steps, troubleshooting, and VRAM shortage solutions for 50x faster local LLM inference
Ollama GPU Scheduling and Resource Management: VRAM Optimization, Multi-GPU Load Balancing
Deep dive into Ollama GPU scheduling and resource management, covering VRAM optimization parameters, multi-GPU load balancing architecture, and llama.cpp technical principles. Includes 3 real-world scenarios to help you run large models stably and maximize multi-GPU hardware
Ollama Performance Optimization: Complete Guide to Quantization, Batch Processing, and Memory Tuning
A deep dive into Ollama quantization techniques (Q4/Q5/Q8 selection strategies), batch processing num_batch configuration for 50-150% throughput improvement, GPU memory management, and OOM solutions. Includes performance benchmarks across different hardware.
Ollama Multi-Model Deployment: Running Qwen, Llama, and DeepSeek in Parallel
A detailed guide on configuring Ollama for multi-model parallel execution, comparing Qwen, Llama, and DeepSeek models, and sharing GPU memory management tips for building an intelligent model switching system.
Ollama + Open WebUI: Build Your Own Local ChatGPT Interface (Complete Guide)
Step-by-step guide to setting up a ChatGPT-style AI interface locally with Ollama and Open WebUI. Covers installation, model selection, RAG knowledge base, API integration, and performance tuning. Get your local AI assistant running in 30 minutes.
Ollama API Calls: From curl to OpenAI SDK Compatible Interface
Learn two ways to call Ollama API: native REST API (curl) and OpenAI SDK compatible interface. Includes complete code examples, streaming response handling, and best practices guide
Ollama API Practice: Python and Node.js Client Development Guide
A comprehensive guide to Ollama API integration, covering Python and Node.js SDK usage, streaming response handling, tool calling with Agent Loop, thinking mode, and OpenAI compatibility comparison
LangChain + Ollama Integration Guide: Complete Local LLM App Development
Complete guide to integrating LangChain with Ollama, with code examples for Chat, RAG, and Agent scenarios, plus OpenAI vs Ollama switching strategies for building enterprise LLM apps with local models.
Ollama Embedding in Practice: Local Vector Search and RAG Setup
Build a local RAG system with Ollama: mxbai-embed-large vs nomic-embed-text model comparison, ChromaDB/FAISS/Milvus vector database selection, complete Python code tutorial
Ollama Model Quantization Guide: GGUF Format and Accuracy Loss Analysis
Deep dive into Ollama GGUF quantization principles, referencing Red Hat's 500K+ evaluation data to reveal accuracy loss truths. Practical quantization recommendations for different hardware configurations to run large models on consumer GPUs.
Ollama Production Monitoring: Logging Configuration and Prometheus Alerting in Practice
A complete Ollama production deployment monitoring solution, including logging configuration, Prometheus metrics collection, AlertManager rules, and Grafana dashboard setup for multi-GPU monitoring and automated fault recovery
Mnemo: A Portable Long-Term Memory Layer for Local LLMs
Mnemo is a local-first AI memory layer. This guide explains how it differs from RAG, how its Rust, SQLite, and graph retrieval architecture works, how to try it with Docker and Ollama, and where its limits are.
