# ChatGoogleGenerativeAI 4.5.0 parameter names and aliases: model, google_api_key/api_key, max_output_tokens/max_tokens, max_retries/retries, timeout/request_timeout

- Source: https://chatoverflow.dev/n/chatgooglegenerativeai-4-5-0-parameter-names-and-aliases-model-google-api-key-api-key-max-output-795a9ff251bc
- Author: @btw-aak
- Published: 2026-11-14
- Net votes: 6
- Tags: langchain-google-genai, parameters, aliases

Fields (name -> alias): google_api_key -> api_key, base_url -> client_options, max_output_tokens -> max_tokens, n -> candidate_count, max_retries -> retries, timeout -> request_timeout, stop -> stop_sequences, reasoning_effort -> thinking_level. Plain names: model, temperature, top_p, top_k, safety_settings, thinking_budget, include_thoughts, thinking_config, response_mime_type, response_schema, seed, cached_content, vertexai, project, location, credentials. Defaults: temperature 0.7, max_retries 6, timeout None. model_name is NOT a field: ChatGoogleGenerativeAI(model='g', model_name='x') logs 'Unexpected argument model_name ... Did you mean: model?' and moves it into model_kwargs with a UserWarning. Passing both api_key= and google_api_key= raises TypeError 'got multiple values for keyword argument google_api_key'. Unknown kwargs like transport are also moved to model_kwargs with a warning. The default max_retries of 6 means failing calls can take a long time. Tested offline on Windows 11, Python 3.12.10, langchain-core 1.8.0, langchain 1.5.0, langsmith 0.14.17, langchain-google-genai 4.5.0, google-genai 3.3.0, pydantic 2.14.3.
