Medical GraphRAG Assistant

# NVIDIA NIM Service Configuration # NIM LLM Configuration nim_llm: model: meta/llama-3.1-8b-instruct container_registry: nvcr.io/nim/meta/llama-3.1-8b-instruct tag: latest port: 8001 # External port (maps to container port 8000) gpu_allocation: all shared_memory: 16g environment: NIM_MODEL_PROFILE: auto health_check: endpoint: /v1/models timeout: 300 # seconds interval: 10 # seconds # NIM Embeddings Configuration (Cloud API) nim_embeddings: model: nvidia/nv-embedqa-e5-v5 api_endpoint: https://api.nvcf.nvidia.com/v2/nvcf/pexec/functions/ embedding_dimension: 1024 batch_size: 50 rate_limit: requests_per_minute: 60 retry_attempts: 3 retry_backoff: exponential # exponential or linear # NIM Vision Configuration nim_vision: model: nvidia/nv-clip-vit container_registry: nvcr.io/nim/nvidia/nv-clip-vit tag: latest port: 8002 # External port (maps to container port 8000) gpu_allocation: all shared_memory: 8g environment: NIM_MODEL_PROFILE: auto health_check: endpoint: /health timeout: 180 # seconds interval: 10 # seconds # Common Docker Settings docker: restart_policy: unless-stopped network_mode: bridge log_driver: json-file log_options: max-size: 100m max-file: "3"