llm 22
- Unreasonable things I ask AI to do
- Planning a four-GPU expansion box for my AM5 workstation
- Combining GPT-6 Astra intelligence with local AI grunt work
- My personal LLM toolchain: switch between models and harnesses, control them from your phone
- Recipe for running Qwen3.8-Flash-Next in NVFP4 on 2x RTX Pro 6000
- Reviving a 30-year-old video game magazine with LLMs
- Forecasting with Foundation Models: Capacity Planning and Incident Detection
- Running DeepSeek-V4-Flash at 700 tokens/s on 2x RTX Pro 6000
- Coding locally with Pi Coding agent and open weights models (April 2026 edition)
- What Happened When I Gave Luna Access to My Email
- Luna: An AI Assistant That Works While I Sleep
- Fixing RTX Pro 6000 Blackwell shutdowns with custom fan control
- The age of hyper-personalized software
- Running MiniMax-M2.1 Locally with Claude Code on Dual RTX Pro 6000
- Guide on installing and running the best models on a dual RTX Pro 6000 rig with vLLM
- Injecting Knowledge into LLMs via Fine-Tuning
- Three Years of ChatGPT
- Getting Started with running LLM models locally
- Silicon Valley's New Secret: Chinese Base Models
- Speeding up local LLM inference 2x with Speculative Decoding
- Open Weights, Borrowed GPUs
- Harnessing GPT-OSS Built-in Tools