Usability benchmark for local LLMs on AMD hardware (RDNA4 + Strix Halo) — agentic throughput, self-repair counts and context-depth decay, all parsed from llama.cpp server logs
-
Updated
Sep 15, 2026 - JavaScript
Usability benchmark for local LLMs on AMD hardware (RDNA4 + Strix Halo) — agentic throughput, self-repair counts and context-depth decay, all parsed from llama.cpp server logs
To associate your repository with the rdna4 topic, visit your repo's landing page and select "manage topics."