From a4f23b64d3bf2d7e1d7ea7f25c44eeb3cefce4cc Mon Sep 17 00:00:00 2001
From: TheCrazyInsanity <inventricks06@gmail.com>
Date: Fri, 8 May 2026 15:48:23 -0400
Subject: [PATCH] idkvro

---
 machines/homepc/configuration.nix  | 15 +++++----------
 modules/virtualization/default.nix |  1 +
 2 files changed, 6 insertions(+), 10 deletions(-)

diff --git a/machines/homepc/configuration.nix b/machines/homepc/configuration.nix
index a542063..0d79fab 100755
--- a/machines/homepc/configuration.nix
+++ b/machines/homepc/configuration.nix
@@ -49,6 +49,7 @@
     pkgs.llama-cpp
     pkgs.llama-swap
     # pkgs.agent-browser
+    pkgs.nvidia-container-toolkit
 
     inputs.llm-agents.packages.${pkgs.stdenv.hostPlatform.system}.pi
     inputs.llm-agents.packages.${pkgs.stdenv.hostPlatform.system}.rtk
@@ -112,20 +113,14 @@
       "Qwen3-4B-Claude-Opus-Distill":
         cmd: llama-server --port ''${PORT} -hf TeichAI/Qwen3-4B-Thinking-2507-Claude-4.5-Opus-High-Reasoning-Distill-GGUF:Q4_K_M --ctx-size 32768 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256
         ttl: 300
-      "Qwen3.5-9B-Thinking":
-        cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-9B-GGUF:Q4_K_M --ctx-size 128000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256 --chat-template-kwargs '{\"enable_thinking\": true}' --no-mmproj --no-kv-offload"
+      "Qwen3.5-9B":
+        cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-9B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
         ttl: 300
       "Qwen3.5-9B-Claude-Opus-Distill":
         cmd: "llama-server --port ''${PORT} -hf Jackrong/Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled-GGUF:Q4_K_S --ctx-size 32000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 512 --chat-template-kwargs '{\"enable_thinking\": true}' --no-mmproj"
         ttl: 300
-      "Qwen3.5-4B-Thinking":
-        cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256 --chat-template-kwargs '{\"enable_thinking\": true}'"
-        ttl: 300
-      "Qwen3.5-9B-Non-Thinking":
-        cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-9B-GGUF:Q4_K_M --ctx-size 32000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256 --no-mmproj"
-        ttl: 300
-      "Qwen3.5-4B-Non-Thinking":
-        cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
+      "Qwen3.5-4B":
+        cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
         ttl: 300
       "Gemma4-E4B":
         cmd: "llama-server --port ''${PORT} -hf Abhiray/gemma-4-E4B-it-heretic-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
diff --git a/modules/virtualization/default.nix b/modules/virtualization/default.nix
index f7c826c..7d64261 100755
--- a/modules/virtualization/default.nix
+++ b/modules/virtualization/default.nix
@@ -14,5 +14,6 @@
   virtualisation.podman = {
     enable = true;
     dockerCompat = true;
+    enableNvidia = true;
   };
 }