nsbradley88 /
talus-public
Rust LLM inference engine tuned for AMD Strix Halo (gfx1151): a fork of Strata with ggml/llama.cpp-derived HIP kernels. Five models on one 128 GB machine.
Updated 49 min ago
Why now
Version v0.1.1 came out on October 31.
+30 stars gained this week
Stars gained each day, last 30 days
Oct 4Oct 19Today
This week Before
In short
It needs an AMD Strix Halo machine (Ryzen AI Max+ 395, Radeon 8060S) with 128 GB of unified memory, and it serves models through an OpenAI-compatible HTTP API. Benchmarks and quality gates are in docs/RESULTS.md.
Summary written by AI from the repository’s description and README.
nsbradley88 / talus-public
View on GitHub(opens in new tab)