mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-08 22:21:35 +00:00
1. Fix test_create_simple_string_input (Gemini and LiteLLM bridge):
- Update usage token checks to support both old format (input_tokens/output_tokens)
and new format (total_input_tokens/total_output_tokens)
- The Interactions API returns total_input_tokens/total_output_tokens
2. Fix test_proxy_function_calling_support_consistency for groq/gemma-7b-it:
- Change expected result from True to False
- Model is not in model_prices_and_context_window.json, so supports_function_calling returns False
- If the model should support function calling, it needs to be added to the JSON file
|
||
|---|---|---|
| .. | ||
| base_interactions_test.py | ||
| test_gemini_interactions.py | ||
| test_google_interactions_integration.py | ||
| test_litellm_responses_bridge.py | ||
| test_openapi_compliance.py | ||