Upstream 1.8.4/1.8.5 are iGPU decode and returning-session speedups with no
packaging change: setup.py still reads EXL3_ROCM_DEV_INCLUDE/EXL3_HIP_DEFINES,
the pyproject runtime deps are unchanged, and tools/{glm,mimo,qwen}/serve.py
are still the serve entry points. Only the pin and source hash move.
Verified on gfx1151: nix build .#kyojin, importing exllamav3_ext against the
pinned ROCm torch 2.11, kyojin-serve-qwen --help, and nix build
.#kyojin.tests.jit.
This commit is contained in:
@@ -18,13 +18,13 @@ let
|
||||
# only supports AMD Strix Halo, so the kernels are built for gfx1151 alone.
|
||||
rocmArch = "gfx1151";
|
||||
|
||||
version = "1.7.1";
|
||||
version = "1.8.5";
|
||||
|
||||
src = fetchFromGitHub {
|
||||
owner = "Yamz-Labs";
|
||||
repo = "kyojin";
|
||||
rev = "v${version}";
|
||||
hash = "sha256-84lRbN7H5eNG/RG8Ygvo9cByGOVOp7moFi0GS/YF+mU=";
|
||||
hash = "sha256-OZhzRnQUrkJOrn2lVatvVxD6JSwtJkwfWwpRUQHawac=";
|
||||
};
|
||||
|
||||
# setup.py passes EXL3_ROCM_DEV_INCLUDE as -I to every translation unit. The
|
||||
|
||||
Reference in New Issue
Block a user