Hegghammer

Working Clawdbot/Moltbot setup with local Ollama model

[Update 2026-02-02: nemotron-3-nano also performs well on same setup; see comment below]

This is a guide to setting up Clawdbot/Moltbot with a local Ollama model that actually works -- meaning it has good tool use and decent speed. The main requirement is 48GB of VRAM. I have yet to find a model that fits on less than this and still works on Moltbot.

The setup involves creating a tool-tuned variant of qwen2.5:72b and modifying a range of configs in Moltbot. At the end you'll get a local Moltbot instance that can use tools (exec, read, write, web search), read skills, and perform agentic tasks without any cloud API dependencies. On my system I get ~16 t/s and have yet to come across a tool/skill that my bot can't use.

Claude Opus wrote the first draft of this Gist, then I (a human) checked and edited it.

maskati

This PowerShell script streams Application Insights live metric traces to your local system and displays them in a PowerShell grid view. This is useful since the Azure Portal trace viewing experience is somewhat limited.

Warning

This script uses undocumented functionality. Exercise caution when running any scripts against your infrastructure.

To use:

1. Ensure you are logged in using Azure CLI and have selected the relevant subscription.
2. Run the script in a PowerShell terminal (tested only with PowerShell 7.3).
3. Select the relevant Application Insights resource from the first grid view. This lists all Application Insights resources in the Azure CL

chrisl8888

if (!document.querySelectorAll) {
  document.querySelectorAll = function (selectors) {
    var style = document.createElement('style'), elements = [], element;
    document.documentElement.firstChild.appendChild(style);
    document._qsa = [];

    style.styleSheet.cssText = selectors + '{x-qsa:expression(document._qsa && document._qsa.push(this))}';
    window.scrollBy(0, 0);
    style.parentNode.removeChild(style);