OpenCode setup with Local models

Notes on this, will publish more as I go: opencode.json { “$schema”: “https://opencode.ai/config.json”, “provider”: { “llama.cpp”: { “npm”: “@ai-sdk/openai-compatible”, “name”: “llama.cpp (local)”, “options”: { “baseURL”: “http://localhost:8000/v1” }, “models”: { “unsloth/Qwen3.5-27B-GGUF:Q6_K”: { “name”: “unsloth/Qwen3.5-27B-GGUF:Q6_K”, “tool_call”: true, “reasoning”: true, “limit”: { “context”: 262144, “output”: 16384 }, “modalities”: { “input”: [ “text”, “image” ], “output”: [ “text” ]…

ROCm with two different GPUs? How hard can it be?

I spent a couple of days this week trying to get vllm to work with more than one GPU.. not the easy way, but two different architecture GPU’s. One gfx1100 (Radeon 7900 XTX) and one gfx908 (Radeon Instinct mi100) – that’s one RDNA3 and one CDNA GPU. Yeah, I’m pushing water uphill here. So now,…

Cooling the mi100 in a desktop case

I’ve had this AMD Instinct mi100 sitting around for waay too long unused.. I tried a couple of different cooling options a while ago, but neither worked satisfactorily, and so it went back on the shelf pending a bit more free time. Of course, “free time” is a myth, I should be more honest and…

Overprovisioning SSD’s on FreeNAS for RAIDZ operation

Part of my new build has me running 5x Samsung 840 Pro 512GB SSD’s as a RAIDZ array (stripes with one drive worth of redundancy), which I’ll be using as a pool for iSCSI targets off which will hang multiple VM’s as their boot drives.  Overkill?  Yes, certainly for a home server. Having just installed…

Fedora Logo

Fedora login manager with multiple screen setup

I’m experimenting with making a switch to Linux for my main desktop, mainly driven by wanting Linux as a development environment for python, rather than Windows. After installing Fedora 21 Workstation version, its become annoying that the login manager, gdm, isn’t remembering the display configuration for my multiple-monitor setup.  I run Portrait Landscape Portrait setup…