M.I.M.I.R - Multi-agent Intelligent Memory & Insight Repository

#!/bin/bash # Build and publish llama.cpp server for ARM64 with mxbai-embed-large (1024 dimensions) set -e # Configuration IMAGE_NAME="timothyswt/llama-cpp-server-arm64-mxbai" VERSION="${1:-latest}" DOCKER_USERNAME="${DOCKER_USERNAME:-timothyswt}" echo "🔨 Building llama.cpp server for ARM64 with mxbai-embed-large..." echo "Image: $IMAGE_NAME:$VERSION" echo "Dimensions: 1024" echo "" # Find and copy model from Ollama models directory OLLAMA_MODELS="$HOME/.ollama/models" MANIFEST_PATH="$OLLAMA_MODELS/manifests/registry.ollama.ai/library/mxbai-embed-large/latest" TARGET_DIR="docker/llama-cpp/models" TARGET_PATH="$TARGET_DIR/mxbai-embed-large.gguf" echo "🔍 Looking for mxbai-embed-large in Ollama models..." if [ ! -f "$MANIFEST_PATH" ]; then echo "❌ Model not found in Ollama" echo "" echo "Please pull the model first:" echo " ollama pull mxbai-embed-large" echo "" exit 1 fi echo "✅ Found Ollama manifest" # Extract model blob digest MODEL_DIGEST=$(jq -r '.layers[] | select(.mediaType == "application/vnd.ollama.image.model") | .digest' "$MANIFEST_PATH" | head -1) if [ -z "$MODEL_DIGEST" ]; then echo "❌ Could not find model digest in manifest" exit 1 fi echo "📦 Model digest: $MODEL_DIGEST" # Copy model blob to target location (Ollama uses dash instead of colon) BLOB_FILE=$(echo "$MODEL_DIGEST" | tr ':' '-') BLOB_PATH="$OLLAMA_MODELS/blobs/$BLOB_FILE" if [ ! -f "$BLOB_PATH" ]; then echo "❌ Model blob not found: $BLOB_PATH" exit 1 fi echo "📋 Copying model to build context..." mkdir -p "$TARGET_DIR" cp "$BLOB_PATH" "$TARGET_PATH" echo "✅ Model copied to: $TARGET_PATH" MODEL_SIZE=$(du -h "$TARGET_PATH" | cut -f1) echo " Size: $MODEL_SIZE" echo "" # Build the image docker build \ --platform linux/arm64 \ -t "$IMAGE_NAME:$VERSION" \ -t "$IMAGE_NAME:latest" \ -f docker/Dockerfile.mxbai \ . echo "" echo "✅ Build complete!" echo "" echo "🔍 Image details:" docker images | grep llama-cpp-server-arm64-mxbai | head -n 2 echo "" echo "📦 Pushing to Docker Hub..." echo " (Make sure you're logged in: docker login)" echo "" # Ask for confirmation read -p "Push to Docker Hub? (y/N) " -n 1 -r echo if [[ $REPLY =~ ^[Yy]$ ]]; then docker push "$IMAGE_NAME:$VERSION" if [ "$VERSION" != "latest" ]; then docker push "$IMAGE_NAME:latest" fi echo "✅ Published to Docker Hub!" else echo "⏭️ Skipped push. To push manually:" echo " docker push $IMAGE_NAME:$VERSION" echo " docker push $IMAGE_NAME:latest" fi # Clean up copied model echo "" echo "🧹 Cleaning up..." rm -f "$TARGET_PATH" echo "✅ Removed temporary model copy" echo "" echo "🎉 Done! To use this image in docker-compose.arm64.yml:" echo " llama-server:" echo " image: $IMAGE_NAME:$VERSION" echo " # ... rest of config ..." echo "" echo " And update environment:" echo " - MIMIR_EMBEDDINGS_MODEL=mxbai-embed-large" echo " - MIMIR_EMBEDDINGS_DIMENSIONS=1024"