llama-cpp: gemma 4 E4B -> gemma 4 E2B
All checks were successful
Build and Deploy / deploy (push) Successful in 2m5s
All checks were successful
Build and Deploy / deploy (push) Successful in 2m5s
This commit is contained in:
@@ -9,8 +9,8 @@
|
|||||||
}:
|
}:
|
||||||
let
|
let
|
||||||
cfg = config.services.llama-cpp;
|
cfg = config.services.llama-cpp;
|
||||||
modelUrl = "https://huggingface.co/unsloth/gemma-4-E4B-it-GGUF/resolve/main/gemma-4-E4B-it-Q4_K_M.gguf";
|
modelUrl = "https://huggingface.co/bartowski/google_gemma-4-E2B-it-GGUF/resolve/main/google_gemma-4-E2B-it-Q4_K_M.gguf";
|
||||||
modelAlias = lib.removeSuffix ".gguf" (builtins.baseNameOf modelUrl);
|
modelAlias = lib.removeSuffix ".gguf" (baseNameOf modelUrl);
|
||||||
in
|
in
|
||||||
{
|
{
|
||||||
services.llama-cpp = {
|
services.llama-cpp = {
|
||||||
@@ -18,7 +18,7 @@ in
|
|||||||
model = toString (
|
model = toString (
|
||||||
pkgs.fetchurl {
|
pkgs.fetchurl {
|
||||||
url = modelUrl;
|
url = modelUrl;
|
||||||
sha256 = "sha256-4bxEJwn+eAqksuybIsFqf83/VC8X8B7Q4yAxFNKPnzQ=";
|
sha256 = "5efe645db4e1909c7a1f4a9608df18e6c14383f5e86777fc49f769f9ba7d5fdf";
|
||||||
}
|
}
|
||||||
);
|
);
|
||||||
port = service_configs.ports.private.llama_cpp.port;
|
port = service_configs.ports.private.llama_cpp.port;
|
||||||
|
|||||||
Reference in New Issue
Block a user