diff --git a/.github/config/morphik.toml b/.github/config/morphik.toml
new file mode 100644
index 00000000..f540f88b
--- /dev/null
+++ b/.github/config/morphik.toml
@@ -0,0 +1,88 @@
+[api]
+host = "0.0.0.0"
+port = 8000
+reload = true
+
+[auth]
+jwt_algorithm = "HS256"
+dev_mode = true
+dev_entity_id = "dev_user"
+dev_entity_type = "developer"
+dev_permissions = ["read", "write", "admin"]
+
+[registered_models]
+openai_gpt4 = { model_name = "gpt-4" }
+openai_gpt4-1 = { model_name = "gpt-4.1" }
+openai_gpt4o = { model_name = "gpt-4o" }
+openai_embedding = { model_name = "text-embedding-3-small" }
+openai_embedding_large = { model_name = "text-embedding-3-large" }
+
+[agent]
+model = "openai_gpt4-1"
+
+[completion]
+model = "openai_gpt4o"
+default_max_tokens = "1000"
+default_temperature = 0.5
+
+[document_analysis]
+model = "openai_gpt4-1"
+
+[database]
+provider = "postgres"
+pool_size = 10
+max_overflow = 15
+pool_recycle = 3600
+pool_timeout = 10
+pool_pre_ping = true
+max_retries = 3
+retry_delay = 1.0
+
+[embedding]
+model = "openai_embedding"
+dimensions = 1536
+similarity_metric = "cosine"
+
+[parser]
+chunk_size = 6000
+chunk_overlap = 300
+use_unstructured_api = false
+use_contextual_chunking = false
+contextual_chunking_model = "openai_gpt4-1"
+
+[parser.vision]
+model = "openai_gpt4-1"
+frame_sample_rate = -1
+
+[reranker]
+use_reranker = true
+provider = "flag"
+model_name = "BAAI/bge-reranker-large"
+query_max_length = 256
+passage_max_length = 512
+use_fp16 = true
+device = "cpu"
+
+[storage]
+provider = "local"
+storage_path = "./storage"
+
+[vector_store]
+provider = "pgvector"
+
+[rules]
+model = "openai_gpt4-1"
+batch_size = 4096
+
+[morphik]
+enable_colpali = true
+mode = "self_hosted"
+api_domain = "api.morphik.ai"
+
+[redis]
+host = "redis"
+port = 6379
+
+[graph]
+model = "openai_gpt4-1"
+enable_entity_resolution = true 
\ No newline at end of file
diff --git a/.github/workflows/docker-build.yml b/.github/workflows/docker-build.yml
index 3e932555..1672487b 100644
--- a/.github/workflows/docker-build.yml
+++ b/.github/workflows/docker-build.yml
@@ -14,6 +14,13 @@ jobs:
       - name: Checkout repository
         uses: actions/checkout@v4
           
+      - name: Copy Docker config
+        run: |
+          echo "Using centralized Docker config from .github/config/morphik.toml"
+          cp .github/config/morphik.toml morphik.toml
+          echo "Config file updated:"
+          head -10 morphik.toml
+
       - name: Free up disk space
         run: |
           sudo rm -rf /usr/share/dotnet
@@ -55,160 +62,130 @@ jobs:
           # Use the local test tag instead of the registry tag
           IMAGE_TAG="morphik-core:test"
           echo "Testing image: $IMAGE_TAG"
-          
-          # Create a minimal config file for testing
-          cat > morphik.toml.test << 'EOF'
-          [api]
-          host = "0.0.0.0"
-          port = 8000
-          reload = true
-          
-          [auth]
-          jwt_algorithm = "HS256"
-          dev_mode = true  # Enabled by default for easier local development
-          dev_entity_id = "dev_user"  # Default dev user ID
-          dev_entity_type = "developer"  # Default dev entity type
-          dev_permissions = ["read", "write", "admin"]  # Default dev permissions
-          
-          #### Registered models
-          [registered_models]
-          
-          # OpenAI models
-          openai_gpt4 = { model_name = "gpt-4" }
-          openai_gpt4-1 = { model_name = "gpt-4.1" }
-          openai_gpt4o = { model_name = "gpt-4o" }
-          
-          # Embedding models
-          openai_embedding = { model_name = "text-embedding-3-small" }
-          openai_embedding_large = { model_name = "text-embedding-3-large" }
-          
-          #### Component configurations ####
-          
-          [agent]
-          model = "openai_gpt4-1"
-          
-          [completion]
-          model = "openai_gpt4o"
-          default_max_tokens = "1000"
-          default_temperature = 0.5
-          
-          [document_analysis]
-          model = "openai_gpt4-1"
-          
-          [database]
-          provider = "postgres"
-          # Connection pool settings
-          pool_size = 10           # Maximum number of connections in the pool
-          max_overflow = 15        # Maximum number of connections that can be created beyond pool_size
-          pool_recycle = 3600      # Time in seconds after which a connection is recycled (1 hour)
-          pool_timeout = 10        # Seconds to wait for a connection from the pool
-          pool_pre_ping = true     # Check connection viability before using it from the pool
-          max_retries = 3          # Number of retries for database operations
-          retry_delay = 1.0        # Initial delay between retries in seconds
-          
-          [embedding]
-          model = "openai_embedding"  # Reference to registered model
-          dimensions = 1536
-          similarity_metric = "cosine"
-          
-          [parser]
-          chunk_size = 6000
-          chunk_overlap = 300
-          use_unstructured_api = false
-          use_contextual_chunking = false
-          contextual_chunking_model = "openai_gpt4-1"  # Reference to a key in registered_models
-          
-          [parser.vision]
-          model = "openai_gpt4-1"  # Reference to a key in registered_models
-          frame_sample_rate = -1  # Set to -1 to disable frame captioning
-          
-          [reranker]
-          use_reranker = true
-          provider = "flag"
-          model_name = "BAAI/bge-reranker-large"
-          query_max_length = 256
-          passage_max_length = 512
-          use_fp16 = true
-          device = "cpu" # use "cpu" if on docker and using a mac, "cuda" if cuda enabled device
-          
-          [storage]
-          provider = "local"
-          storage_path = "./storage"
-          
-          # [storage]
-          # provider = "aws-s3"
-          # region = "us-east-2"
-          # bucket_name = "morphik-s3-storage"
-          
-          [vector_store]
-          provider = "pgvector"
-          
-          [rules]
-          model = "openai_gpt4-1"
-          batch_size = 4096
-          
-          [morphik]
-          enable_colpali = true
-          mode = "self_hosted"  # "cloud" or "self_hosted"
-          api_domain = "api.morphik.ai"  # API domain for cloud URIs
-          
-          [redis]
-          host = "redis"  # use "redis" for docker
-          port = 6379
-          
-          [graph]
-          model = "openai_gpt4-1"
-          enable_entity_resolution = true
-          
-          # Start container in detached mode with config mounted
-          CONTAINER_ID=$(docker run -d -p 8000:8000 \
-            -e POSTGRES_URI="postgresql://morphik:morphik@localhost:5432/morphik" \
-            -v "$(pwd)/morphik.toml.test:/app/morphik.toml" \
+
+          # Create a custom network for the containers to communicate
+          NETWORK_NAME="morphik-test-network"
+          docker network create $NETWORK_NAME
+
+          # Start pgvector PostgreSQL container
+          echo "Starting pgvector PostgreSQL container..."
+          DB_CONTAINER_ID=$(docker run -d \
+            --name pgvector-test \
+            --network $NETWORK_NAME \
+            -e POSTGRES_USER=morphik \
+            -e POSTGRES_PASSWORD=morphik \
+            -e POSTGRES_DB=morphik \
+            -p 5432:5432 \
+            ankane/pgvector:latest)
+
+          # Wait for PostgreSQL to be ready
+          echo "Waiting for PostgreSQL to be ready..."
+          for i in {1..30}; do
+            if docker exec $DB_CONTAINER_ID pg_isready -U morphik -d morphik; then
+              echo "PostgreSQL is ready!"
+              break
+            fi
+            if [ $i -eq 30 ]; then
+              echo "PostgreSQL failed to start"
+              docker logs $DB_CONTAINER_ID
+              exit 1
+            fi
+            sleep 1
+          done
+
+          # Enable pgvector extension
+          echo "Enabling pgvector extension..."
+          docker exec $DB_CONTAINER_ID psql -U morphik -d morphik -c 'CREATE EXTENSION IF NOT EXISTS vector;'
+
+          # Start Redis container (NOTE: must be named "redis")
+          echo "Starting Redis container..."
+          REDIS_CONTAINER_ID=$(docker run -d \
+            --name redis \
+            --network $NETWORK_NAME \
+            -p 6379:6379 \
+            -v redis_data:/data \
+            redis:7-alpine \
+            redis-server --appendonly yes)
+
+          # Wait for Redis to be ready
+          echo "Waiting for Redis to be ready..."
+          for i in {1..30}; do
+            if docker exec $REDIS_CONTAINER_ID redis-cli ping | grep -q PONG; then
+              echo "Redis is ready!"
+              break
+            fi
+            if [ $i -eq 30 ]; then
+              echo "Redis failed to start"
+              docker logs $REDIS_CONTAINER_ID
+              exit 1
+            fi
+            sleep 1
+          done
+
+          # Start application container
+          echo "Starting application container..."
+          CONTAINER_ID=$(docker run -d \
+            --network $NETWORK_NAME \
+            -p 8000:8000 \
+            -e POSTGRES_URI="postgresql+asyncpg://morphik:morphik@pgvector-test:5432/morphik" \
+            -v "$(pwd)/morphik.toml:/app/morphik.toml" \
             "$IMAGE_TAG")
-          
+
           echo "Started container: $CONTAINER_ID"
-          
-          # Wait for server to be ready with 60 second timeout
-          timeout=60
+
+          # Wait for server to be ready
+          timeout=300  # long timeout required to load checkpoint shards
           interval=2
           elapsed=0
-          
+
           echo "Waiting for server to be ready..."
           while [ $elapsed -lt $timeout ]; do
             if curl -f -s http://localhost:8000/ping > /dev/null 2>&1; then
               echo "✅ Server is responding to /ping endpoint"
               break
             fi
-            
+
             echo "⏳ Waiting for server... (${elapsed}s/${timeout}s)"
             sleep $interval
             elapsed=$((elapsed + interval))
           done
-          
+
           # Check if we timed out
           if [ $elapsed -ge $timeout ]; then
             echo "❌ Server failed to respond within ${timeout} seconds"
             echo "Container logs:"
             docker logs "$CONTAINER_ID"
-            docker stop "$CONTAINER_ID"
-            docker rm "$CONTAINER_ID"
+            echo "Database logs:"
+            docker logs "$DB_CONTAINER_ID"
+            echo "Redis logs:"
+            docker logs "$REDIS_CONTAINER_ID"
+            docker stop "$CONTAINER_ID" "$DB_CONTAINER_ID" "$REDIS_CONTAINER_ID" || true
+            docker rm -f "$CONTAINER_ID" "$DB_CONTAINER_ID" "$REDIS_CONTAINER_ID" || true
+            docker network rm "$NETWORK_NAME" || true
+            docker volume rm redis_data 2>/dev/null || true
             exit 1
           fi
-          
+
           # Verify the response is actually 200
-          HTTP_CODE=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8000/ping)
-          if [ "$HTTP_CODE" = "200" ]; then
-            echo "✅ Health check passed - /ping returned HTTP $HTTP_CODE"
-          else
-            echo "❌ Health check failed - /ping returned HTTP $HTTP_CODE"
+          response_code=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8000/ping)
+          if [ "$response_code" -ne 200 ]; then
+            echo "❌ Unexpected response code: $response_code"
+            echo "Container logs:"
             docker logs "$CONTAINER_ID"
-            docker stop "$CONTAINER_ID"
-            docker rm "$CONTAINER_ID"
+            echo "Database logs:"
+            docker logs "$DB_CONTAINER_ID"
+            echo "Redis logs:"
+            docker logs "$REDIS_CONTAINER_ID"
+            docker stop "$CONTAINER_ID" "$DB_CONTAINER_ID" "$REDIS_CONTAINER_ID" || true
+            docker rm -f "$CONTAINER_ID" "$DB_CONTAINER_ID" "$REDIS_CONTAINER_ID" || true
+            docker network rm "$NETWORK_NAME" || true
+            docker volume rm redis_data 2>/dev/null || true
             exit 1
           fi
-          
-          # Clean up
-          echo "🧹 Cleaning up container"
-          docker stop "$CONTAINER_ID"
-          docker rm "$CONTAINER_ID"
-          echo "✅ Test completed successfully" 
\ No newline at end of file
+
+          echo "✅ Tests passed!"
+          docker stop "$CONTAINER_ID" "$DB_CONTAINER_ID" "$REDIS_CONTAINER_ID" || true
+          docker rm -f "$CONTAINER_ID" "$DB_CONTAINER_ID" "$REDIS_CONTAINER_ID" || true
+          docker network rm "$NETWORK_NAME" || true
+          docker volume rm redis_data 2>/dev/null || true
+          echo "✅ Test completed successfully"
\ No newline at end of file
diff --git a/.github/workflows/docker-publish.yml b/.github/workflows/docker-publish.yml
index ed7f12ed..416ef627 100644
--- a/.github/workflows/docker-publish.yml
+++ b/.github/workflows/docker-publish.yml
@@ -4,6 +4,8 @@ on:
   push:
     branches:
       - main
+      # we publish our own config docker image from this branch
+      - feat/issue-158-publish-docker-image
       
 permissions:
   contents: read
@@ -34,6 +36,13 @@ jobs:
           sudo docker builder prune -a --force
           df -h
 
+      - name: Copy Docker config
+        run: |
+          echo "Using centralized Docker config from .github/config/morphik.toml"
+          cp .github/config/morphik.toml morphik.toml
+          echo "Config file updated:"
+          head -10 morphik.toml
+
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
         
diff --git a/docker-compose.yml b/docker-compose.yml
index beb6fb96..31c662a0 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -40,9 +40,8 @@ services:
         condition: service_healthy
       config-check:
         condition: service_completed_successfully
-      ollama:
-        condition: service_started
-        required: false
+      # ollama:
+      #   condition: service_started
     networks:
       - morphik-network
     env_file:
@@ -70,9 +69,8 @@ services:
         condition: service_healthy
       config-check:
         condition: service_completed_successfully
-      ollama:
-        condition: service_started
-        required: false
+      # ollama:
+      #   condition: service_started
     networks:
       - morphik-network
     env_file:
diff --git a/docker/docker-entrypoint.sh b/docker/docker-entrypoint.sh
new file mode 100755
index 00000000..b4969ece
--- /dev/null
+++ b/docker/docker-entrypoint.sh
@@ -0,0 +1,58 @@
+#!/bin/bash
+set -e
+
+# Copy default config if none exists
+if [ ! -f /app/morphik.toml ]; then
+    cp /app/morphik.toml.default /app/morphik.toml
+fi
+
+# Function to check PostgreSQL
+check_postgres() {
+    if [ -n "$POSTGRES_URI" ]; then
+        # Extract connection details from POSTGRES_URI, which can be
+        # postgresql:// or postgresql+asyncpg://
+
+        # Using awk for more robust URI parsing that handles special characters
+        eval $(./parse-postgres-uri.py "$POSTGRES_URI")
+
+        echo "Waiting for PostgreSQL..."
+        max_retries=30
+        retries=0
+        until PGPASSWORD=$PG_PASS pg_isready -h $PG_HOST -p $PG_PORT -U $PG_USER -d $PG_DB; do
+            retries=$((retries + 1))
+            if [ $retries -eq $max_retries ]; then
+                echo "Error: PostgreSQL did not become ready in time"
+                exit 1
+            fi
+            echo "Waiting for PostgreSQL... (Attempt $retries/$max_retries)"
+            sleep 2
+        done
+        echo "PostgreSQL is ready!"
+        
+        # Verify database connection
+        # NOTE: preserve stderr for debugging
+        if ! PGPASSWORD=$PG_PASS psql -h $PG_HOST -p $PG_PORT -U $PG_USER -d $PG_DB -c "SELECT 1"; then
+            echo "Error: Could not connect to PostgreSQL database"
+            echo "POSTGRES_URI: $POSTGRES_URI"
+            echo "USER: $PG_USER"
+            echo "PASS: $PG_PASS"
+            echo "HOST: $PG_HOST"
+            echo "PORT: $PG_PORT"
+            echo "DB: $PG_DB"
+            exit 1
+        fi
+        echo "PostgreSQL connection verified!"
+    fi
+}
+
+# Check PostgreSQL
+check_postgres
+
+# Check if command arguments were passed ($# is the number of arguments)
+if [ $# -gt 0 ]; then
+    # If arguments exist, execute them (e.g., execute "arq core.workers...")
+    exec "$@"
+else
+    # Otherwise, execute the default command (uv run start_server.py)
+    exec uv run uvicorn core.api:app --host $HOST --port $PORT --loop asyncio --http auto --ws auto --lifespan auto
+fi
\ No newline at end of file
diff --git a/docker/morphik.toml.default b/docker/morphik.toml.default
new file mode 100644
index 00000000..6c7b24ee
--- /dev/null
+++ b/docker/morphik.toml.default
@@ -0,0 +1,41 @@
+[api]
+host = "0.0.0.0"
+port = 8000
+reload = false
+
+[auth]
+jwt_algorithm = "HS256"
+dev_mode = true
+dev_entity_id = "dev_user"
+dev_entity_type = "developer"
+dev_permissions = ["read", "write", "admin"]
+
+[completion]
+provider = "ollama"
+model_name = "llama2"
+base_url = "http://localhost:11434"
+
+[database]
+provider = "postgres"
+
+[embedding]
+provider = "ollama"
+model_name = "nomic-embed-text"
+dimensions = 768
+similarity_metric = "cosine"
+base_url = "http://localhost:11434"
+
+[parser]
+chunk_size = 1000
+chunk_overlap = 200
+use_unstructured_api = false
+
+[reranker]
+use_reranker = false
+
+[storage]
+provider = "local"
+storage_path = "/app/storage"
+
+[vector_store]
+provider = "pgvector"
diff --git a/docker/parse-postgres-uri.py b/docker/parse-postgres-uri.py
new file mode 100755
index 00000000..29f9a39a
--- /dev/null
+++ b/docker/parse-postgres-uri.py
@@ -0,0 +1,102 @@
+#!/usr/bin/env python3
+import re
+import sys
+import urllib.parse
+from typing import Dict
+
+def parse_postgres_uri(uri: str) -> Dict[str, str]:
+    """Parse a PostgreSQL connection URI into its components.
+    
+    Args:
+        uri: The PostgreSQL connection URI (e.g., 'postgresql://user:pass@host:port/db')
+        
+    Returns:
+        Dictionary containing the parsed components (USER, PASS, HOST, PORT, DB)
+    """
+    # Default values
+    result = {
+        'USER': '',
+        'PASS': '',
+        'HOST': '',
+        'PORT': '5432',  # Default PostgreSQL port
+        'DB': 'postgres'  # Default database name
+    }
+    
+    try:
+        # PostgreSQL URI pattern:
+        # postgresql[+driver]://[user[:password]@][netloc][:port][/dbname][?param1=value1&...]
+        pattern = r'''
+            ^
+            postgresql(?:\+[a-z]+)?://  # scheme with optional driver
+            (?:([^:/?#@]+)(?::([^/?#@]*))?@)?  # user:password@
+            (?:([^:/?#]+)(?::(\d+))?)?  # host:port
+            (?:/([^?#]*))?  # /dbname
+            (?:\?([^#]*))?  # ?query
+            $
+        '''
+        
+        match = re.match(pattern, uri.strip(), re.VERBOSE)
+        if not match:
+            raise ValueError("Invalid PostgreSQL URI format")
+            
+        user, password, host, port, dbname, query = match.groups()
+        
+        # Handle username and password
+        if user:
+            result['USER'] = urllib.parse.unquote(user)
+        if password:
+            result['PASS'] = urllib.parse.unquote(password)
+        
+        # Handle host and port
+        if host:
+            result['HOST'] = host
+        if port:
+            result['PORT'] = port
+        
+        # Handle database name
+        if dbname:
+            result['DB'] = urllib.parse.unquote(dbname)
+        
+        # Handle query parameters (e.g., for password in query string)
+        if query and not result['PASS']:
+            for param in query.split('&'):  # type: ignore
+                if '=' in param:
+                    key, value = param.split('=', 1)
+                    if key.lower() == 'password':
+                        result['PASS'] = urllib.parse.unquote(value)
+                        break
+    
+    except Exception as e:
+        # If any error occurs, print the failure message and exit
+        print(f"Error parsing PostgreSQL URI: {e}", file=sys.stderr)
+        print("POSTGRES_URI_PARSE_FAILURE=1", file=sys.stderr)
+        sys.exit(1)
+    
+    return result
+
+def main():
+    if len(sys.argv) != 2:
+        print("Usage: eval $(parse-postgres-uri.sh \"postgresql://user:pass@host:port/db\")", file=sys.stderr)
+        sys.exit(1)
+    
+    # Print default empty values first (for compatibility with original script)
+    print('PG_USER=""')
+    print('PG_PASS=""')
+    print('PG_HOST=""')
+    print('PG_PORT=""')
+    print('PG_DB=""')
+    
+    # Parse the URI and print the results
+    try:
+        components = parse_postgres_uri(sys.argv[1])
+        for key, value in components.items():
+            # Escape special characters in the value for shell compatibility
+            escaped_value = value.replace('"', '\\"').replace('`', '\\`').replace('$', '\\$')
+            print(f'PG_{key}="{escaped_value}"')
+    except Exception as e:
+        print(f'Error parsing PostgreSQL URI: {e}', file=sys.stderr)
+        print('POSTGRES_URI_PARSE_FAILURE=1')
+        sys.exit(1)
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/dockerfile b/dockerfile
index 0d3c9d67..39f5a97c 100644
--- a/dockerfile
+++ b/dockerfile
@@ -100,97 +100,12 @@ ENV PORT=8000
 ENV VIRTUAL_ENV=/app/.venv
 ENV PATH="/app/.venv/bin:/usr/local/bin:${PATH}"
 
-# Create default configuration
-RUN echo '[api]\n\
-host = "0.0.0.0"\n\
-port = 8000\n\
-reload = false\n\
-\n\
-[auth]\n\
-jwt_algorithm = "HS256"\n\
-dev_mode = true\n\
-dev_entity_id = "dev_user"\n\
-dev_entity_type = "developer"\n\
-dev_permissions = ["read", "write", "admin"]\n\
-\n\
-[completion]\n\
-provider = "ollama"\n\
-model_name = "llama2"\n\
-base_url = "http://localhost:11434"\n\
-\n\
-[database]\n\
-provider = "postgres"\n\
-\n\
-[embedding]\n\
-provider = "ollama"\n\
-model_name = "nomic-embed-text"\n\
-dimensions = 768\n\
-similarity_metric = "cosine"\n\
-base_url = "http://localhost:11434"\n\
-\n\
-[parser]\n\
-chunk_size = 1000\n\
-chunk_overlap = 200\n\
-use_unstructured_api = false\n\
-\n\
-[reranker]\n\
-use_reranker = false\n\
-\n\
-[storage]\n\
-provider = "local"\n\
-storage_path = "/app/storage"\n\
-\n\
-[vector_store]\n\
-provider = "pgvector"\n\
-' > /app/morphik.toml.default
+# Copy our custom configuration as the default
+COPY morphik.toml /app/morphik.toml.default
 
 # Create startup script
-RUN echo '#!/bin/bash\n\
-set -e\n\
-\n\
-# Copy default config if none exists\n\
-if [ ! -f /app/morphik.toml ]; then\n\
-    cp /app/morphik.toml.default /app/morphik.toml\n\
-fi\n\
-\n\
-# Function to check PostgreSQL\n\
-check_postgres() {\n\
-    if [ -n "$POSTGRES_URI" ]; then\n\
-        echo "Waiting for PostgreSQL..."\n\
-        max_retries=30\n\
-        retries=0\n\
-        until PGPASSWORD=$PGPASSWORD pg_isready -h postgres -U morphik -d morphik; do\n\
-            retries=$((retries + 1))\n\
-            if [ $retries -eq $max_retries ]; then\n\
-                echo "Error: PostgreSQL did not become ready in time"\n\
-                exit 1\n\
-            fi\n\
-            echo "Waiting for PostgreSQL... (Attempt $retries/$max_retries)"\n\
-            sleep 2\n\
-        done\n\
-        echo "PostgreSQL is ready!"\n\
-        \n\
-        # Verify database connection\n\
-        if ! PGPASSWORD=$PGPASSWORD psql -h postgres -U morphik -d morphik -c "SELECT 1" > /dev/null 2>&1; then\n\
-            echo "Error: Could not connect to PostgreSQL database"\n\
-            exit 1\n\
-        fi\n\
-        echo "PostgreSQL connection verified!"\n\
-    fi\n\
-}\n\
-\n\
-# Check PostgreSQL\n\
-check_postgres\n\
-\n\
-# Check if command arguments were passed ($# is the number of arguments)\n\
-if [ $# -gt 0 ]; then\n\
-    # If arguments exist, execute them (e.g., execute "arq core.workers...")\n\
-    exec "$@"\n\
-else\n\
-    # Otherwise, execute the default command (uv run start_server.py)\n\
-    exec uv run uvicorn core.api:app --host $HOST --port $PORT --loop asyncio --http auto --ws auto --lifespan auto\n\
-fi\n\
-' > /app/docker-entrypoint.sh && chmod +x /app/docker-entrypoint.sh
+COPY docker/parse-postgres-uri.py /app/parse-postgres-uri.py
+COPY docker/docker-entrypoint.sh /app/docker-entrypoint.sh
 
 # Copy application code
 # pyproject.toml is needed for uv to identify the project context for `uv run`
diff --git a/morphik.toml b/morphik.toml
index 231370c7..895bb4b1 100644
--- a/morphik.toml
+++ b/morphik.toml
@@ -13,8 +13,9 @@ dev_permissions = ["read", "write", "admin"]  # Default dev permissions
 #### Registered models
 [registered_models]
 # OpenAI models
+openai_gpt4 = { model_name = "gpt-4" }
 openai_gpt4-1 = { model_name = "gpt-4.1" }
-openai_gpt4-1-mini = { model_name = "gpt-4.1-mini" }
+openai_gpt4o = { model_name = "gpt-4o" }
 
 # Azure OpenAI models
 azure_gpt4 = { model_name = "gpt-4", api_base = "YOUR_AZURE_URL_HERE", api_version = "2023-05-15", deployment_id = "gpt-4-deployment" }
@@ -32,21 +33,24 @@ ollama_qwen_vision = { model_name = "ollama_chat/qwen2.5vl:latest", api_base = "
 ollama_llama_vision = { model_name = "ollama_chat/llama3.2-vision", api_base = "http://localhost:11434", vision = true }
 ollama_embedding = { model_name = "ollama/nomic-embed-text", api_base = "http://localhost:11434" }
 
+# Embedding models
 openai_embedding = { model_name = "text-embedding-3-small" }
 openai_embedding_large = { model_name = "text-embedding-3-large" }
 azure_embedding = { model_name = "text-embedding-ada-002", api_base = "YOUR_AZURE_URL_HERE", api_version = "2023-05-15", deployment_id = "embedding-ada-002" }
 
-
 #### Component configurations ####
 
 [agent]
-model = "ollama_qwen_vision" # Model for the agent logic
+model = "openai_gpt4-1"
 
 [completion]
-model = "ollama_qwen_vision" #"openai_gpt4-1-mini"  # Reference to a key in registered_models
+model = "openai_gpt4o"
 default_max_tokens = "1000"
 default_temperature = 0.5
 
+[document_analysis]
+model = "openai_gpt4-1"
+
 [database]
 provider = "postgres"
 # Connection pool settings
@@ -59,8 +63,8 @@ max_retries = 3          # Number of retries for database operations
 retry_delay = 1.0        # Initial delay between retries in seconds
 
 [embedding]
-model = "ollama_embedding"  # Reference to registered model
-dimensions = 768
+model = "openai_embedding"  # Reference to registered model
+dimensions = 1536
 similarity_metric = "cosine"
 
 [parser]
@@ -68,13 +72,10 @@ chunk_size = 6000
 chunk_overlap = 300
 use_unstructured_api = false
 use_contextual_chunking = false
-contextual_chunking_model = "ollama_qwen_vision"  # Reference to a key in registered_models
-
-[document_analysis]
-model = "ollama_qwen_vision"  # Reference to a key in registered_models
+contextual_chunking_model = "openai_gpt4-1"  # Reference to a key in registered_models
 
 [parser.vision]
-model = "ollama_qwen_vision"  # Reference to a key in registered_models
+model = "openai_gpt4-1"  # Reference to a key in registered_models
 frame_sample_rate = -1  # Set to -1 to disable frame captioning
 
 [reranker]
@@ -84,7 +85,7 @@ model_name = "BAAI/bge-reranker-large"
 query_max_length = 256
 passage_max_length = 512
 use_fp16 = true
-device = "mps" # use "cpu" if on docker and using a mac, "cuda" if cuda enabled device
+device = "cpu" # use "cpu" if on docker and using a mac, "cuda" if cuda enabled device
 
 [storage]
 provider = "local"
@@ -99,7 +100,7 @@ storage_path = "./storage"
 provider = "pgvector"
 
 [rules]
-model = "ollama_qwen_vision"
+model = "openai_gpt4-1"
 batch_size = 4096
 
 [morphik]
@@ -111,11 +112,11 @@ morphik_embedding_api_domain = "http://localhost:6000"  # endpoint for multivect
 colpali_mode = "local" # "off", "local", or "api"
 
 [redis]
-host = "localhost"  # use "redis" for docker
+host = "redis"  # use "redis" for docker
 port = 6379
 
 [graph]
-model = "ollama_qwen_vision"
+model = "openai_gpt4-1"
 enable_entity_resolution = true
 
 [telemetry]