llama-cpp: update + re-enable + gemma 4 E4B

This commit is contained in:
2026-04-03 14:06:35 -04:00
parent e765a98487
commit d4d01d63f1
2 changed files with 6 additions and 6 deletions

6
flake.lock generated
View File

@@ -302,11 +302,11 @@
]
},
"locked": {
"lastModified": 1775161985,
"narHash": "sha256-CahR+F3hF7C5IYWooVblvlzHf2eDST0Eagu59xdY37g=",
"lastModified": 1775236905,
"narHash": "sha256-tHshzR/k6D/r5UhJCfJ9b/mJgsbn7ODtnZrDlimhOOI=",
"owner": "TheTom",
"repo": "llama-cpp-turboquant",
"rev": "63b832bc0799ba7270e695e0987d0bd2272bdc7e",
"rev": "bc05a6803e48f17e0f2c7a99fce9b50d03882de7",
"type": "github"
},
"original": {

View File

@@ -8,11 +8,11 @@
}:
{
services.llama-cpp = {
enable = false;
enable = true;
model = toString (
pkgs.fetchurl {
url = "https://huggingface.co/unsloth/Qwen3.5-9B-GGUF/resolve/main/Qwen3.5-9B-Q4_K_M.gguf";
sha256 = "03b74727a860a56338e042c4420bb3f04b2fec5734175f4cb9fa853daf52b7e8";
url = "https://huggingface.co/unsloth/gemma-4-E4B-it-GGUF/resolve/main/gemma-4-E4B-it-Q4_K_M.gguf";
sha256 = "ced37f54b80068fe65e95c6dd79ac88cddc227e179fd1040b8f751b1e5bdf849";
}
);
port = service_configs.ports.private.llama_cpp.port;