fix: upgrade to Q6_K (600MB) better quality

Dockerfile

@@ -10,8 +10,8 @@ RUN pip install --no-cache-dir --timeout 300 llama-cpp-python==0.3.23 \
 RUN apt-get update && apt-get install -y --no-install-recommends curl \
     && rm -rf /var/lib/apt/lists/*
 RUN mkdir -p /app/models && \
-    curl -sL -o /app/models/MiniCPM-V-4_6-Thinking-Q4_K_M.gguf \
-    "https://huggingface.co/openbmb/MiniCPM-V-4.6-Thinking-gguf/resolve/main/MiniCPM-V-4_6-Thinking-Q4_K_M.gguf" && \
+    curl -sL -o /app/models/MiniCPM-V-4_6-Thinking-Q6_K.gguf \
+    "https://huggingface.co/openbmb/MiniCPM-V-4.6-Thinking-gguf/resolve/main/MiniCPM-V-4_6-Thinking-Q6_K.gguf" && \
     curl -sL -o /app/models/mmproj-model-f16.gguf \
     "https://huggingface.co/openbmb/MiniCPM-V-4.6-Thinking-gguf/resolve/main/mmproj-model-f16.gguf"
 

app.py

@@ -11,7 +11,7 @@ _ready = False
 _llm = None
 _llm_lock = threading.Lock()
 _tasks = {}
-MODEL_PATH = "/app/models/MiniCPM-V-4_6-Thinking-Q4_K_M.gguf"
+MODEL_PATH = "/app/models/MiniCPM-V-4_6-Thinking-Q6_K.gguf"
 MMPROJ_PATH = "/app/models/mmproj-model-f16.gguf"