aasheeshlikepanner/cascadeflow

Set these on your host before deploying. Each platform has an “Environment Variables” section in the deploy flow.

• OPENAI_API_KEY secret OpenAI Industry-leading quality, most reliable, best for production
• OPENAI_TOOL_MODEL secret Get your key: https://platform.openai.com/api-keys Models: gpt-4, gpt-4-turbo, gpt-4o, gpt-4o-mini, gpt-3.5-turbo Optional: used by live…
• ANTHROPIC_API_KEY secret Anthropic Claude Best for reasoning and analysis, strong safety features
• GROQ_API_KEY secret Groq Fastest inference speed, ultra-low latency, free tier available
• TOGETHER_API_KEY secret Together AI Cost-effective, wide model selection, good for experimentation
• HF_TOKEN secret Hugging Face Open-source models, community-driven, flexible deployment
• HF_INFERENCE_ENDPOINT_URL Optional: If using Hugging Face Inference Endpoints
• GOOGLE_API_KEY secret Google (Vertex AI) Enterprise integration, GCP ecosystem, Gemini models
• AZURE_API_KEY secret Azure OpenAI Enterprise compliance, HIPAA/SOC2, Microsoft ecosystem
• AZURE_OPENAI_ENDPOINT secret
• AZURE_OPENAI_DEPLOYMENT_NAME secret
• AZURE_OPENAI_API_VERSION secret
• DEEPSEEK_API_KEY secret DeepSeek Specialized code models, very cost-effective for coding tasks
• OPENROUTER_API_KEY secret OpenRouter Unified API for 400+ models from multiple providers
• OLLAMA_BASE_URL For local Ollama (default)
• VLLM_BASE_URL For local vLLM (default)
• VLLM_MODEL_NAME Optional: vLLM model configuration
• LOG_LEVEL Logging level (DEBUG, INFO, WARNING, ERROR)
• CASCADEFLOW_DEFAULT_QUALITY_THRESHOLD Default quality threshold for responses (0.0 1.0)