SurfSense/surfsense_backend/app/config/global_llm_config.example.yaml
Ojārs Kapteinis f8c5c5c96e Remove Google Analytics and Anthropic Claude from default configs
Privacy and configuration improvements:
- Remove Google Analytics tracking from frontend (privacy enhancement)
- Remove Anthropic Claude 3 Opus from example LLM configurations
- Renumber example config IDs sequentially (-1 through -4)

Technical details:
- Removed @next/third-parties/google import from layout.tsx
- Removed GoogleAnalytics component from HTML root
- Removed Claude example from global_llm_config.example.yaml
- Production uses three-tier local-first architecture (Mistral NeMo + TildeOpen + Gemini fallback)

These changes align with the local-first European AI architecture already
deployed at https://ai.kapteinis.lv using Mistral NeMo 12B (France) and
TildeOpen 30B (Latvia) with Gemini as emergency fallback only.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 21:42:57 +02:00

68 lines
2 KiB
YAML

# Global LLM Configuration
#
# SETUP INSTRUCTIONS:
# 1. For production: Copy this file to global_llm_config.yaml and add your real API keys
# 2. For testing: The system will use this example file automatically if global_llm_config.yaml doesn't exist
#
# NOTE: The example API keys below are placeholders and won't work.
# Replace them with your actual API keys to enable global configurations.
#
# These configurations will be available to all users as a convenient option
# Users can choose to use these global configs or add their own
global_llm_configs:
# Example: OpenAI GPT-4 Turbo
- id: -1
name: "Global GPT-4 Turbo"
provider: "OPENAI"
model_name: "gpt-4-turbo-preview"
api_key: "sk-your-openai-api-key-here"
api_base: ""
language: "English"
litellm_params:
temperature: 0.7
max_tokens: 4000
# Example: Fast model - GPT-3.5 Turbo
- id: -2
name: "Global GPT-3.5 Turbo"
provider: "OPENAI"
model_name: "gpt-3.5-turbo"
api_key: "sk-your-openai-api-key-here"
api_base: ""
language: "English"
litellm_params:
temperature: 0.5
max_tokens: 2000
# Example: Chinese LLM - DeepSeek
- id: -3
name: "Global DeepSeek Chat"
provider: "DEEPSEEK"
model_name: "deepseek-chat"
api_key: "your-deepseek-api-key-here"
api_base: "https://api.deepseek.com/v1"
language: "Chinese"
litellm_params:
temperature: 0.7
max_tokens: 4000
# Example: Groq - Fast inference
- id: -4
name: "Global Groq Llama 3"
provider: "GROQ"
model_name: "llama3-70b-8192"
api_key: "your-groq-api-key-here"
api_base: ""
language: "English"
litellm_params:
temperature: 0.7
max_tokens: 8000
# Notes:
# - Use negative IDs to distinguish global configs from user configs
# - IDs should be unique and sequential (e.g., -1, -2, -3, etc.)
# - The 'api_key' field will not be exposed to users via API
# - Users can select these configs for their long_context, fast, or strategic LLM roles
# - All standard LiteLLM providers are supported