Local, ternary-weight LLM inference on AMD Strix Halo. Rust above the kernels, HIP below, zero Python at runtime. https://discord.gg/EhQgmNePg
rust amd mcp hip rocm inference-engine 1-bit bitnet npu openai-api llm llm-inference local-ai strix-halo gfx1151 1-58-bit ternary-llm xdna2
-
Updated
May 2, 2026 - HTML