servius/dotfiles
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

feat(chat): implement qwen3-coder-30b model with custom chat template and tool support

Browse Source 
The commit message reflects the main changes:
1. Updated model name from "gpt-oss-20b" to "qwen3-coder-30b" in aichat.nix
2. Changed the default model to ryu:qwen3-coder-30b in aichat.nix
3. Added a new chat template file (chat.hbs) with tool/function calling support
4. Configured llama.cpp to use the new chat template file
5. Enabled HDR display settings in hyprland
6. Modified gamemode to unload lmstudio on start and added custom script for gamemode startup

All changes are focused on implementing Qwen3-Coder-30B model with enhanced tool calling capabilities and system configuration improvements.
This commit is contained in:
uttarayan21
2025-11-26 15:34:10 +05:30
parent 3a8e182170
commit 58594d582b
5 changed files with 144 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
home/programs/aichat.nix
View File

@@ -50,7 +50,7 @@
api_base = "https://llama.ryu.darksailor.dev/v1";
models = [
{
name = "gpt-oss-20b";
name = "qwen3-coder-30b";
type = "chat";
}
# {
@@ -173,7 +173,7 @@
*/
''
---
model: openai:gpt-4o
model: ryu:qwen3-coder-30b
---
Your task is to generate a concise and informative commit message based on the provided diff. Use the conventional commit format, which includes a type (feat, fix, chore, docs, style, refactor, perf, test) and an optional scope. The message should be in the imperative mood and should not exceed 72 characters in the subject line. Don't include any additional text or explanations, just the commit message.
'';