This post walks through how to configure Clawdbot to utilize Docker Model Runner, enabling a privacy-first approach to personal intelligence.<\/p>\n
![\"Clawdbot](\"https:\/\/www.docker.com\/app\/uploads\/2026\/01\/Clawdbot-figure-1.png\" "\\"-")
\n <\/div>\nWhat Are Clawdbot and Docker Model Runner?<\/h2>\n
Clawdbot<\/strong> is a self-hosted AI assistant designed to live where you already are. Unlike browser-bound bots, Clawdbot integrates directly with messaging apps like Telegram, WhatsApp, Discord, and Signal<\/strong>. It acts as a proactive digital coworker capable of executing real-world actions across your devices and services.<\/p>\n Docker Model Runner (DMR)<\/strong> is Docker\u2019s native solution for running and managing large language models (LLMs) as OCI artifacts. It exposes an OpenAI-compatible API, allowing it to serve as the private \u201cbrain\u201d for any tool that supports standard AI endpoints.<\/p>\n Together, they create a unified assistant that can browse the web, manage your files, and respond to your messages without ever sending your sensitive data to a third-party cloud.<\/p>\n Privacy by Design<\/strong><\/p>\n In a \u201cPrivacy-First\u201d setup, your assistant\u2019s memory, message history, and files stay on your hardware. Docker Model Runner isolates model inference, meaning:<\/p>\n Cost Control and Scaling<\/strong><\/p>\n Cloud-based agents often become expensive when they use \u201clong-term memory\u201d or \u201cproactive searching,\u201d which consume massive amounts of tokens. With Docker Model Runner, inference runs on your own GPU\/CPU. Once a model is pulled, there are no per-token fees<\/strong>. You can let Clawdbot summarize thousands of unread emails or research complex topics for hours without worrying about a surprise API bill at the end of the month.<\/p>\n Clawdbot uses a flexible configuration system to define which models and providers drive its reasoning. While the onboarding wizard (clawdbot onboard) is the standard setup path, you can manually point Clawdbot to your private Docker infrastructure.<\/p>\n You can define your provider configuration in:<\/p>\n To bridge the two, update your configuration to point to the DMR server. Assuming Docker Model Runner is running at its default address: http:\/\/localhost:12434\/v1.<\/p>\n Your config.json should be updated as follows:<\/p>\n This configuration tells Clawdbot to bypass external APIs and route all \u201cthinking\u201d to your private models.<\/p>\n\n Note for Docker Desktop Users:<\/strong> While coding models focus on logic, personal assistant models need a balance of instruction-following, tool-use capability, and long-term memory.<\/p>\nBenefits of the Clawdbot + DMR Stack<\/h2>\n
\n
\n
\n
Configuring Clawdbot with Docker Model Runner<\/h2>\n
Modifying the Clawdbot Configuration<\/strong><\/h3>\n
\n
Using Clawdbot with Docker Model Runner<\/h3>\n
\n{\n \"models\": {\n \"providers\": {\n \"dmr\": {\n \"baseUrl\": \"http:\/\/localhost:12434\/v1\",\n \"apiKey\": \"dmr-local\",\n \"api\": \"openai-completions\",\n \"models\": [\n {\n \"id\": \"gpt-oss:128K\",\n \"name\": \"gpt-oss (128K context window)\",\n \"contextWindow\": 128000,\n \"maxTokens\": 128000\n },\n {\n \"id\": \"glm-4.7-flash:128K\",\n \"name\": \"glm-4.7-flash (128K context window)\",\n \"contextWindow\": 128000,\n \"maxTokens\": 128000\n }\n ]\n }\n }\n },\n \"agents\": {\n \"defaults\": {\n \"model\": {\n \"primary\": \"dmr\/gpt-oss:128K\"\n }\n }\n }\n}\n\n<\/pre>\n<\/div>\n
Ensure TCP access is enabled so Clawdbot can communicate with the runner. Run the following command in your terminal:
docker desktop enable model-runner \u2013tcp<\/p>\n<\/div>\nRecommended Models for Personal Assistants<\/h2>\n