Category: AI
-
Building a 6-bit LLM decoder for AMD XDNA1: what actually got faster
From failed per-operation offload and broken 4-bit quality to a working 24-layer W6 decoder, real GPU comparisons, and a CPU/NPU/GPU experiment.
-
oc-box: More Freedom for AI Coding, Now With a Dashboard
oc-box grows into a more approachable OpenCode workspace: a terminal dashboard, encrypted backups, clipboard attachments and simpler upgrades.