Llm on smcleod.net

Patching NVIDIA's driver and vLLM to enable P2P on consumer GPUs

Wed, 25 Feb 2026 23:59:00 +1000

NVIDIA artificially restricts peer-to-peer (P2P) GPU communication to their enterprise cards. Turns out this is a software limitation, not a hardware one. I patched my drivers to remove it, hacked vLLM to take advantage of it, and got a 15-50% throughput improvement running Qwen 3.5 35b on dual RTX 3090s.

The Role Bridging Problem

Thu, 20 Nov 2025 01:10:00 +1000

An observation on functional correctness without domain quality.

Stop Polluting Context - Let Users Disable Individual MCP Tools

Tue, 26 Aug 2025 01:10:00 +1000

If you’re building MCP servers, you should be adding the ability to disable individual tools.

MCP DevTools

Mon, 18 Aug 2025 01:10:00 +1000

A single, modular MCP server for AI coding agents.

Agentic Coding Adoption Cost Cycle

Wed, 16 Jul 2025 01:10:00 +1000

The two common themes I see with engineers adopting agentic coding tools

Agentic Coding Workflow & Cline Demo

Fri, 20 Jun 2025 01:10:00 +1000

Recording of a demo and Q&A session on my Agentic Coding workflow and Cline use.

Vibe Coding vs Agentic Coding

Fri, 06 Jun 2025 01:10:00 +1000

From Creative Exploration to Production Quality

My Plan, Document, Act, Review flow for Agentic Software Development

Mon, 28 Apr 2025 01:10:00 +1000

An overview of my agentic coding development flow.

LLM Sampling Parameters Guide

Fri, 25 Apr 2025 01:10:00 +1000

A practical guide to LLM sampling parameters for Ollama, llama.cpp, and MLX

Getting Started with Agentic Systems - Developer Learning Paths

Wed, 16 Apr 2025 01:10:00 +1000

A curated learning path for engineers looking to gain practical experience with AI and agentic systems.

The Cost of Agentic Coding

Thu, 03 Apr 2025 01:00:01 +1000

The cost of not leveraging agentic coding is likely far greater than you think.

The Democratisation Paradox: What History Teaches Us About AI

Sun, 23 Mar 2025 01:00:01 +1000

The Democratisation Paradox: What History Teaches Us About AI.

The effects of prompt caching on Agentic coding

Thu, 20 Mar 2025 21:00:01 +1000

What is prompt caching and why is it so important for Agentic coding?

Agentic Coding - Live Demo / Brownbag

Fri, 07 Feb 2025 01:00:01 +1000

Video recording of a brownbag presentation / live demo I ran on Agentic Coding using Cline/Roo Code

Bringing K/V Context Quantisation to Ollama

Wed, 04 Dec 2024 20:00:02 +0000

K/V context cache quantisation has been added to Ollama. This enables significant reductions in VRAM usage, allowing users to realise the potential of expanded context sizes and run larger models at their existing context sizes.

Will AI Take My Job?

Fri, 29 Nov 2024 22:00:02 +0000

It's probably not so much AI itself you have to fear.

Generating Diagrams with with AI / LLMs

Tue, 08 Oct 2024 01:00:10 +0000

Generating diagrams with AI / LLMs

Ingest: Streamlining Content Preparation for LLMs

Mon, 29 Jul 2024 01:00:10 +0000

A CLI tool for parsing directories into LLM-friendly markdown

LLM Parameter Playground

Sat, 20 Jul 2024 01:00:10 +0000

A web app to explore the effects of different inference parameters on LLMs.

Code, Chaos, and Copilots (AI/LLM Talk July 2024)

Thu, 18 Jul 2024 01:00:10 +0000

Integrating AI into Daily Development

Understanding AI/LLM Quantisation Through Interactive Visualisations

Wed, 17 Jul 2024 01:00:10 +0000

AI/LLM Quantisation Visualised

Gollama: Ollama Model Manager

Sat, 01 Jun 2024 01:00:10 +0000

Gollama is a Go-based client for Ollama for managing models.

Confuddlement: Download Confluence Spaces as Markdown, Summarise with Ollama

Thu, 23 May 2024 05:45:10 +0000

Download Confluence Spaces as Markdown, Summarise with Ollama

SuperPrompter - Supercharge your text prompts for AI/LLM image generation

Fri, 22 Mar 2024 00:00:00 +0000

SuperPrompter is a Python-based application that utilises the SuperPrompt-v1 model to generate optimised text prompts for AI/LLM image generation (for use with Stable Diffusion etc…) from user prompts.

See Brian Fitzgerald’s Blog for a detailed explanation of the SuperPrompt-v1 model and its capabilities / limitations.

Features

Utilises the SuperPrompt-v1 model for text generation.
A basic (aka ugly) graphical user interface built with tkinter.
Customisable generation parameters (max new tokens, repetition penalty, temperature, top p, top k, seed).
Optional logging of input parameters and generated outputs.
Bundling options to include or exclude pre-downloaded model files.
Unloads the models when the application is idle to free up memory.

Prebuilt Binaries

Check releases page to see if there are any prebuilt binaries available for your platform.

Llamalink - Ollama to LM Studio LLM Model Linker

Thu, 21 Mar 2024 00:00:00 +0000

This is a simple command-line tool that allows you to easily link Ollama models to LM Studio's directory structure. It simplifies the process of symlinking Ollama models to LM Studio, making it convenient to use the models in both applications.

Open source, locally hosted AI powered Siri replacement

Mon, 20 Nov 2023 01:00:00 +0000

Using an ESP32 S3 Box 3 with Willow, Home Assistant and Large Language Models (LLMs) create a locally hosted, offline, AI powered Siri / Alexa replacement.

Introduction to AI and Large Language Models (LLMs)

Thu, 26 Jan 2023 01:00:00 +0000

A high level intro to LLMs that I'm writing for a few friends that are new to the concept. It is far from complete, definitely contains some errors and is a work in progress.

LLM FAQ

Mon, 01 Jan 0001 00:00:00 +0000

Frequently Asked Questions about LLMs and AI

LLM vRAM Estimator

Mon, 01 Jan 0001 00:00:00 +0000

LLM vRAM Estimator