main/config.yaml

server:
  host: "0.0.0.0"
  port: 8001

model:
  base_path: "."
  generation:
    max_new_tokens: 500
    do_sample: true
    temperature: 0.2
    repetition_penalty: 1.1
  defaults:
    #model_name: "huihui-ai/Llama-3.2-3B-Instruct-abliterated"
    model_name: "huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated"

folders:
  models: "main/models"
  cache: "main/.cache"
  logs: "main/logs"

logging:
  level: "DEBUG" # DEBUG, INFO, WARNING, ERROR, CRITICAL
  format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
  file: "llm_api.log"

api:
  version: "v1"
  prefix: "/api"
  cors:
    origins: ["*"]
    credentials: true