Files
stack/config/router-config.yaml
zeekay a96cdb7d29 rebrand: rename litellm to llm — configs and compose files
litellm_params → llm_params, python -m litellm → python -m llm.
2026-02-25 15:13:22 -08:00

137 lines
3.6 KiB
YAML

model_list:
# OpenAI Models
- model_name: gpt-4o
llm_params:
model: gpt-4o
api_key: os.environ/OPENAI_API_KEY
- model_name: gpt-4o-mini
llm_params:
model: gpt-4o-mini
api_key: os.environ/OPENAI_API_KEY
- model_name: gpt-4-turbo
llm_params:
model: gpt-4-turbo
api_key: os.environ/OPENAI_API_KEY
- model_name: gpt-3.5-turbo
llm_params:
model: gpt-3.5-turbo
api_key: os.environ/OPENAI_API_KEY
# Anthropic Models
- model_name: claude-3-5-sonnet-20241022
llm_params:
model: claude-3-5-sonnet-20241022
api_key: os.environ/ANTHROPIC_API_KEY
- model_name: claude-3-opus-20240229
llm_params:
model: claude-3-opus-20240229
api_key: os.environ/ANTHROPIC_API_KEY
- model_name: claude-3-sonnet-20240229
llm_params:
model: claude-3-sonnet-20240229
api_key: os.environ/ANTHROPIC_API_KEY
- model_name: claude-3-haiku-20240307
llm_params:
model: claude-3-haiku-20240307
api_key: os.environ/ANTHROPIC_API_KEY
- model_name: claude-3-5-haiku-20241022
llm_params:
model: claude-3-5-haiku-20241022
api_key: os.environ/ANTHROPIC_API_KEY
# Google/Gemini Models
- model_name: gemini-pro
llm_params:
model: gemini-pro
api_key: os.environ/GEMINI_API_KEY
- model_name: gemini-pro-vision
llm_params:
model: gemini-pro-vision
api_key: os.environ/GEMINI_API_KEY
# Groq Models
- model_name: llama-3.2-90b-text-preview
llm_params:
model: groq/llama-3.2-90b-text-preview
api_key: os.environ/GROQ_API_KEY
- model_name: llama-3.2-70b-versatile
llm_params:
model: groq/llama-3.2-70b-versatile
api_key: os.environ/GROQ_API_KEY
- model_name: mixtral-8x7b-32768
llm_params:
model: groq/mixtral-8x7b-32768
api_key: os.environ/GROQ_API_KEY
# Together AI Models (if available)
- model_name: mixtral-8x22b-instruct
llm_params:
model: together_ai/mixtral-8x22b-instruct
api_key: os.environ/TOGETHER_API_KEY
# Router Settings with Hanzo Console Observability
router_settings:
routing_strategy: simple-shuffle
model_group_alias:
gpt-4: ["gpt-4o", "gpt-4-turbo"]
claude-3: ["claude-3-opus-20240229", "claude-3-sonnet-20240229", "claude-3-haiku-20240307"]
claude-3.5: ["claude-3-5-sonnet-20241022", "claude-3-5-haiku-20241022"]
gemini: ["gemini-pro", "gemini-pro-vision"]
llama: ["llama-3.2-90b-text-preview", "llama-3.2-70b-versatile"]
drop_params: true
max_retries: 2
request_timeout: 600
telemetry: true
success_callback: ["console"]
failure_callback: ["console"]
set_verbose: false
json_logs: true
# Hanzo Console integration
console_public_key: os.environ/CONSOLE_PUBLIC_KEY
console_secret_key: os.environ/CONSOLE_SECRET_KEY
console_host: os.environ/CONSOLE_HOST
# General Settings
general_settings:
master_key: os.environ/MASTER_KEY
database_url: os.environ/DATABASE_URL
redis_url: os.environ/REDIS_URL
max_parallel_requests: 100
infer_model_from_keys: true
# custom_auth: /app/config/router-extensions.py
# Enable key validation with Cloud
enable_key_based_auth: false
# key_management_system: "cloud"
# MCP Settings
mcp_servers:
hanzo_mcp:
command: uvx
args:
- hanzo-mcp
env:
WORKSPACE_PATH: /workspace
# Proxy routes to other services
proxy_routes:
- pattern: "^/runtime/.*"
target: os.environ/RUNTIME_URL
strip_prefix: "/runtime"
- pattern: "^/services/.*"
target: os.environ/SERVICES_URL
strip_prefix: "/services"
- pattern: "^/chat/.*"
target: os.environ/CHAT_URL
strip_prefix: "/chat"