chore(llm): update llama-cpp, llama-swap and switch to qwen3.5-27b-thinking

- Bump llama-cpp from version 8157 to 8196
- Bump llama-swap from version 192 to 197
- Switch default assistant model from qwen3-coder-next-80b to qwen3.5-27b-thinking
- Remove glm-4-32b-instruct model configuration
- Update qwen3.5-27b-thinking config:
  - Use bartowski quantization (IQ4_XS) instead of unsloth
  - Increase context window from 131k to 196k
  - Add cache type settings (q8_0) and CUDA device
- Add 1password-cli to home-manager programs
- Fix typo: 'dispay' -> 'display' in llm-config.lua
This commit is contained in:
2026-03-05 07:32:57 -05:00
parent 9234782492
commit 1bce17c5f9
5 changed files with 14 additions and 31 deletions

View File

@@ -13,13 +13,13 @@ let
in
buildGoModule (finalAttrs: {
pname = "llama-swap";
version = "192";
version = "197";
src = fetchFromGitHub {
owner = "mostlygeek";
repo = "llama-swap";
tag = "v${finalAttrs.version}";
hash = "sha256-CMzF935cREAFfWHt5yzX05wvp/DC/3GWZZfhRtJVYaA=";
hash = "sha256-EXgyYmpbN/zzr6KeSpvFEB+FS7gDIZFinNMv70v5boY=";
# populate values that require us to use git. By doing this in postFetch we
# can delete .git afterwards and maintain better reproducibility of the src.
leaveDotGit = true;
@@ -35,7 +35,7 @@ buildGoModule (finalAttrs: {
vendorHash = "sha256-XiDYlw/byu8CWvg4KSPC7m8PGCZXtp08Y1velx4BR8U=";
passthru.ui = callPackage ./ui.nix { llama-swap = finalAttrs.finalPackage; };
passthru.npmDepsHash = "sha256-4VH9jJ1Ae16p8kUubZBrIwwqw/X8I+wDg378G82WCtU=";
passthru.npmDepsHash = "sha256-Fs7+JKE8YBp2Xj8bVBlwmT+UwuD642VeUHiPx+fv94c=";
nativeBuildInputs = [
versionCheckHook