Small doc fixes

.env.example

@@ -1,8 +1,11 @@
 # Change the name of this file to .env after updating it!
 
 ############
-# [required]
-# n8n credentials - you set this to whatever you want, just make it a long and secure string for both!
+# [required] 
+# n8n credentials - use the command `openssl rand -hex 32` to generate both
+#   openssl is available by default on Linux/Mac
+#   For Windows, you can use the 'Git Bash' terminal installed with git
+#   Or run the command: python -c "import secrets; print(secrets.token_hex(32))"
 ############
 
 N8N_ENCRYPTION_KEY=super-secret-key

README.md

@@ -235,7 +235,7 @@ to the IP address of your cloud instance.
 
 **NOTE**: If you are using a cloud machine without the "docker compose" command available by default, such as a Ubuntu GPU instance on DigitalOcean, run these commands before running start_services.py:
 
-- DOCKER_COMPOSE_VERSION=$(curl -s https://api.github.com/repos/docker/compose/releases/latest | grep 'tag_name' | cut -d\" -f4)
+- DOCKER_COMPOSE_VERSION=$(curl -s https://api.github.com/repos/docker/compose/releases/latest | grep 'tag_name' | cut -d\\" -f4)
 - sudo curl -L "https://github.com/docker/compose/releases/download/${DOCKER_COMPOSE_VERSION}/docker-compose-linux-x86_64" -o /usr/local/bin/docker-compose
 - sudo chmod +x /usr/local/bin/docker-compose
 - sudo mkdir -p /usr/local/lib/docker/cli-plugins

docker-compose.yml

@@ -37,6 +37,9 @@ x-ollama: &service-ollama
     - 11434/tcp
   environment:
     - OLLAMA_CONTEXT_LENGTH=8192
+    - OLLAMA_FLASH_ATTENTION=1
+    - OLLAMA_KV_CACHE_TYPE=q8_0
+    - OLLAMA_MAX_LOADED_MODELS=2
   volumes:
     - ollama_storage:/root/.ollama
 
@@ -49,10 +52,6 @@ x-init-ollama: &init-ollama
   command:
     - "-c"
     - "sleep 3; OLLAMA_HOST=ollama:11434 ollama pull qwen2.5:7b-instruct-q4_K_M; OLLAMA_HOST=ollama:11434 ollama pull nomic-embed-text"
-    # For a larger context length verison of the model, run these commands:
-    # echo "FROM qwen2.5:7b-instruct-q4_K_M\n\nPARAMETER num_ctx 8096" > Modelfile
-    # ollama create qwen2.5:7b-8k -f ./Modelfile
-    # Change the name of the LLM and num_ctx as you see fit.
 
 services:
   flowise: