fix: Update model file references and enhance Dockerfile for LlamaCPP setup

ai_stack/llamacpp/Dockerfile

@@ -7,6 +7,8 @@ RUN apt-get update && apt-get install -y \
     g++ \
     wget \
     curl \
+    libcurl4-openssl-dev \
+    pkg-config \
     && rm -rf /var/lib/apt/lists/*
 
 # Clone llama.cpp
@@ -16,6 +18,11 @@ RUN git clone https://github.com/ggerganov/llama.cpp.git
 WORKDIR /llama.cpp
 RUN mkdir build && cd build && cmake .. -DLLAMA_SERVER=ON && make -j$(nproc)
 
+# Copy the built executable to a standard location
+RUN cp /llama.cpp/build/bin/llama-server /usr/local/bin/llama-server || \
+    cp /llama.cpp/build/llama-server /usr/local/bin/llama-server || \
+    find /llama.cpp -name "llama-server" -type f -executable -exec cp {} /usr/local/bin/llama-server \;
+
 # Create models directory
 RUN mkdir -p /models
 
@@ -27,4 +34,4 @@ HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
     CMD curl -f http://localhost:8080/health || exit 1
 
 # Default command - will be overridden by docker-compose
-CMD ["/llama.cpp/llama-server", "--host", "0.0.0.0", "--port", "8080"]
+CMD ["llama-server", "--host", "0.0.0.0", "--port", "8080"]

ai_stack/llamacpp/README.md

@@ -173,7 +173,7 @@ ls -la ai_stack/llamacpp/models/
 # Test with minimal configuration
 docker run --rm -v $(pwd)/ai_stack/llamacpp/models:/models \
   llamacpp /llama.cpp/llama-server \
-  --model /models/DeepSeek-R1-0528-Qwen3-8B-Q4_K_XL.gguf \
+  --model /models/DeepSeek-R1-0528-Qwen3-8B-UD-Q4_K_XL.gguf \
   --host 0.0.0.0 --port 8080 --ctx-size 2048
 ```
 

ai_stack/llamacpp/docker-compose.yml

@@ -2,6 +2,7 @@ version: '3.8'
 services:
   llama-cpp-server:
     build: .
+    image: localhost/llama-cpp-server:latest
     ports:
       - "8080:8080" # Map host port 8080 to container port 8080
     volumes:
@@ -12,21 +13,16 @@ services:
       - LLAMA_ARG_PORT=8080
     restart: unless-stopped
     command: [
-      "/llama.cpp/llama-server",
-      "--model", "/models/DeepSeek-R1-0528-Qwen3-8B-Q4_K_XL.gguf",
+      "llama-server",
+      "--model", "/models/DeepSeek-R1-0528-Qwen3-8B-UD-Q4_K_XL.gguf",
       "--host", "0.0.0.0",
       "--port", "8080",
       "--ctx-size", "32768",
-      "--n-predict", "2048",
       "--threads", "4",
       "--batch-size", "512",
       "--cont-batching",
       "--parallel", "2",
-      "--flash-attn",
-      "--temperature", "0.7",
-      "--top-p", "0.9",
-      "--top-k", "40",
-      "--repeat-penalty", "1.1"
+      "--flash-attn"
     ]
     # Uncomment for GPU support
     # deploy:

ai_stack/ollama/docker-compose.yml

@@ -2,6 +2,7 @@ version: '3.8'
 services:
   ollama:
     build: .
+    image: localhost/ollama-server:latest
     ports:
       - "11434:11434" # Map host port 11434 to container port 11434 (Ollama default)
     volumes: