[aiapienthusiast]

This adds Cheaper Inference as a chat provider: ChatCheaperInference.
Cheaper Inference is an OpenAI-compatible LLM gateway.
One API key gives access to models from several labs.
It uses the same pattern as ChatDeepSeek and ChatNvidia: a small ChatOpenAI subclass with its own base URL.
The same provider was added to Windows-Use in Jeomon/Windows-Use#14.

• New module src/providers/cheaper_inference/.
• The default model is gpt-5.4-mini.
• The key comes from CHEAPER_INFERENCE_API_KEY.
• CHEAPER_INFERENCE_API_BASE can override the base URL.
• ChatCheaperInference is exported from src.providers.

Usage and tests

from src.providers.cheaper_inference import ChatCheaperInference

llm = ChatCheaperInference(model="gpt-5.4-mini")

• tests/test_cheaper_inference.py checks the defaults and the base URL override.
  It mocks the OpenAI clients, so it needs no key and no network.

Base URL: https://api.cheaperinference.com/v1
Docs: https://cheaperinference.com/docs