Popular repositories Loading
-
qwen38-256k-on-16gb
qwen38-256k-on-16gb PublicQwen3.8-27B at 256k context on a 16 GB AMD RX 9070 XT (llama.cpp Vulkan): measured, quality-gated, reproducible. 56.6-58.5 t/s at working depth, ~42-46 t/s at 200k resident with a selective-attenti…
Shell 4
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.