86 lines
2.4 KiB
YAML
86 lines
2.4 KiB
YAML
# LiteLLM — sovereign LLM router per v4.0 master stack
|
|
# OpenRouter = universal provider (1 key, all models)
|
|
# Direct DeepSeek + Groq as backups
|
|
#
|
|
# ACTIVE providers with real keys:
|
|
# - openrouter → DeepSeek, MiniMax, Gemini, Mistral, Llama
|
|
# - deepseek → direct
|
|
# - groq → fast inference
|
|
#
|
|
# PLACEHOLDER (no real key, skipped): gemini, zai, minimax, mistral-direct, kimi
|
|
|
|
model_list:
|
|
# PRIMARY: OpenRouter (universal)
|
|
- model_name: deepseek-v3
|
|
litellm_params:
|
|
model: openrouter/deepseek/deepseek-chat
|
|
api_key: os.environ/OPENROUTER_API_KEY
|
|
api_base: https://openrouter.ai/api/v1
|
|
rpm: 60
|
|
timeout: 60
|
|
- model_name: MiniMax-M3
|
|
litellm_params:
|
|
model: openrouter/MiniMax/MiniMax-M3
|
|
api_key: os.environ/OPENROUTER_API_KEY
|
|
api_base: https://openrouter.ai/api/v1
|
|
rpm: 60
|
|
timeout: 60
|
|
- model_name: gemini-2.5-flash
|
|
litellm_params:
|
|
model: openrouter/google/gemini-2.5-flash
|
|
api_key: os.environ/OPENROUTER_API_KEY
|
|
api_base: https://openrouter.ai/api/v1
|
|
rpm: 60
|
|
timeout: 60
|
|
- model_name: mistral-large
|
|
litellm_params:
|
|
model: openrouter/mistralai/mistral-large-latest
|
|
api_key: os.environ/OPENROUTER_API_KEY
|
|
api_base: https://openrouter.ai/api/v1
|
|
rpm: 60
|
|
timeout: 60
|
|
- model_name: llama-3-70b
|
|
litellm_params:
|
|
model: openrouter/meta-llama/llama-3-70b-instruct
|
|
api_key: os.environ/OPENROUTER_API_KEY
|
|
api_base: https://openrouter.ai/api/v1
|
|
rpm: 60
|
|
timeout: 60
|
|
|
|
# BACKUP: direct DeepSeek
|
|
- model_name: deepseek-direct
|
|
litellm_params:
|
|
model: deepseek/deepseek-chat
|
|
api_key: os.environ/DEEPSEEK_API_KEY
|
|
api_base: https://api.deepseek.com/v1
|
|
rpm: 60
|
|
timeout: 30
|
|
|
|
# DIRECT: Groq (fast inference)
|
|
- model_name: groq-llama-3.3-70b
|
|
litellm_params:
|
|
model: groq/llama-3.3-70b-versatile
|
|
api_key: os.environ/GROQ_API_KEY
|
|
rpm: 100
|
|
timeout: 30
|
|
- model_name: groq-llama-3.1-8b
|
|
litellm_params:
|
|
model: groq/llama-3.1-8b-instant
|
|
api_key: os.environ/GROQ_API_KEY
|
|
rpm: 100
|
|
timeout: 30
|
|
|
|
# FALLBACK: local Ollama
|
|
- model_name: ollama-llama
|
|
litellm_params:
|
|
model: ollama/llama3.2
|
|
api_base: http://host.docker.internal:11434
|
|
rpm: 30
|
|
timeout: 60
|
|
|
|
litellm_settings:
|
|
drop_params: true
|
|
set_verbose: false
|
|
request_timeout: 60
|
|
num_retries: 2
|
|
telemetry: false
|