Nov 15, 2026, 12:08 PM

I've got OpenCode using both my LLM servers simultaneously. Doing some updates on my website and adding a bunch of pics of retro projects that were missing. #opencode #agent #agenticai #localai #homelab

A dark-themed desktop screen split between two system monitoring and AI inference dashboards. On the left, the gpuview dashboard displays live hardware telemetry for three NVIDIA GeForce RTX 2080 Ti GPUs on the "solar-ai" node, showing temperatures between 40°C and 50°C, utilization from 15% to 32%, and high memory usage, with a smaller chat or agent window layered in the bottom-left corner running an image analysis workflow. On the right, the Strata monitor interface is shown actively "Reading prompt" for 6,000 tokens, displaying performance metrics including a decode speed of 47.3 tokens per second, 59.2 out of 64 GB of VRAM allocated across four GPUs, an average GPU temperature of 48°C, 157W of power draw, 1% CPU usage, 2% context fill, and a log of recent inference requests at the bottom.