first commit

2026-04-05 15:07:53 +08:00
commit d1bac85f28
30 changed files with 1071 additions and 0 deletions
@@ -0,0 +1,120 @@
+# =============================================================================
+# LiteLLM Proxy Config (DB-backed UI, no Redis)
+# - Single-node deployment (1 CPU / 2 GB RAM)
+# - Postgres used for:
+#     - Users / login
+#     - Virtual keys & teams
+#     - Spend tracking (optional)
+# - No Redis caching (cache: false)
+#
+# Docs:
+#   https://docs.litellm.ai/docs/proxy/configs
+#   https://docs.litellm.ai/docs/proxy/config_settings
+#   https://docs.litellm.ai/docs/proxy/virtual_keys
+# =============================================================================
+
+# -----------------------------------------------------------------------------#
+# Models exposed by the proxy
+# -----------------------------------------------------------------------------#
+model_list:
+  # --- OpenAI examples -------------------------------------------------------
+  #- model_name: gpt-4o
+  #  litellm_params:
+  #    model: openai/gpt-4o
+  #    api_key: os.environ/OPENAI_API_KEY
+
+  #- model_name: gpt-4o-mini
+  #  litellm_params:
+  #    model: openai/gpt-4o-mini
+  #    api_key: os.environ/OPENAI_API_KEY
+
+  # --- Anthropic example -----------------------------------------------------
+  # - model_name: claude-3-5-sonnet
+  #   litellm_params:
+  #     model: anthropic/claude-3-5-sonnet-20241022
+  #     api_key: os.environ/ANTHROPIC_API_KEY
+
+  # --- Groq example ----------------------------------------------------------
+  # - model_name: groq-llama-3.3-70b
+  #   litellm_params:
+  #     model: groq/llama-3.3-70b-versatile
+  #     api_key: os.environ/GROQ_API_KEY
+
+  # --- Azure OpenAI example --------------------------------------------------
+  # - model_name: azure-gpt-4o
+  #   litellm_params:
+  #     model: azure/my_azure_deployment
+  #     api_base: os.environ/AZURE_API_BASE
+  #     api_key: os.environ/AZURE_API_KEY
+  #     api_version: "2025-01-01-preview"
+
+  # --- Local Ollama example --------------------------------------------------
+  # - model_name: ollama-llama3
+  #   litellm_params:
+  #     model: ollama/llama3
+  #     api_base: http://host.docker.internal:11434
+
+# -----------------------------------------------------------------------------#
+# Core LiteLLM behavior (no Redis cache)
+# -----------------------------------------------------------------------------#
+litellm_settings:
+  # Retries & timeouts
+  num_retries: 2
+  request_timeout: 60        # seconds
+
+  # Disable caching entirely to avoid Redis
+  cache: false               # <- IMPORTANT: no Redis in your stack
+
+  # Drop unsupported provider-specific params instead of erroring
+  drop_params: true
+
+  # If you later enable caching, add cache_params here with Redis config.
+  # cache_params:
+  #   type: redis
+  #   host: litellm-redis
+  #   port: 6379
+  #   password: os.environ/REDIS_PASSWORD
+
+# -----------------------------------------------------------------------------#
+# Router settings (all in-memory, single-node)
+# -----------------------------------------------------------------------------#
+router_settings:
+  routing_strategy: "simple-shuffle"   # simple-shuffle | least-busy | latency-based-routing
+  num_retries: 1
+  timeout: 30
+  retry_after: 5
+
+  # No Redis routing / transaction buffer on this small single-node setup
+  # redis_host: litellm-redis
+  # redis_port: 6379
+  # redis_password: os.environ/REDIS_PASSWORD
+
+# -----------------------------------------------------------------------------#
+# General settings (DB-backed UI, no Redis)
+# -----------------------------------------------------------------------------#
+general_settings:
+  # Admin master key (used for UI + admin API).
+  # Must match LITELLM_MASTER_KEY from .env and start with "sk-".
+  master_key: os.environ/LITELLM_MASTER_KEY
+
+  # Postgres connection (from DATABASE_URL env var).
+  # Required for:
+  #   - /ui login and user management
+  #   - virtual keys
+  #   - spend tracking
+  database_url: os.environ/DATABASE_URL
+
+  # Store model definitions in DB (used by the UI)
+  store_model_in_db: true
+
+  # Telemetry & metrics
+  telemetry: false
+  enable_prometheus: true
+
+  # On a small box, you may want to reduce some DB-heavy features.
+  # You can still track spend, but disable some advanced cost mgmt.
+  disable_end_user_cost_tracking: false
+  disable_reset_budget: false
+
+  # No Redis transaction buffer here
+  # use_redis_transaction_buffer: false