config.yml

---
model_url: https://huggingface.co./TheBloke/llama2_7b_chat_uncensored-GGML
typer:
  delay: 0.1
runpod:
  endpoint_id: vxi8yc4mxz187n
  prefer_async: true
llm:
  max_tokens: 600
  top_k:
  top_p:
  temperature:
  repetition_penalty:
  last_n_tokens:
  seed: -1
  batch_size: 8
  threads: -1
  stop:
    - "</s>"
queue:
  max_size: 16
  concurrency_count: 1  # recommend setting this no larger than your current