Answer generation model versions and lifecycle for the Answer API (Legacy)

Important: Gemini Enterprise retrieves search results from the assistant using StreamAssist. We recommend using the StreamAssist instead of the Search and Answer. For more information, see Get search results using StreamAssist.

Gemini Enterprise offers multiple model versions for you to choose when generating answers. You can choose the model versions when using search summaries and answers and follow-ups.

Available models

Gemini Enterprise uses two types of models for question and answering use cases:

• Vertex AI LLM models that have been tested on question and answering tasks
• Gemini Enterprise models that are based on Vertex AI LLM models and further trained to address question and answering tasks

Gemini Enterprise models share the same discontinuation date as their base Vertex AI LLM models. The base LLM model is available for six months after the release date of the next version of the model, per the Vertex AI model lifecycle policy. Leave enough time to migrate to new models before the discontinuation dates.

The following table lists model version specifications. When you set a model specification, the API uses the specified model to generate answers.

Model version	Description	Context window	Discontinuation date
stable	The default model choice if the model version is not set. The stable model specification points to gemini-2.5-flash/answer_gen/v1. The model designated as stable changes periodically as new models and versions become available.	128K	N/A
gemini-2.5-flash/answer_gen/v1	A Gemini Enterprise model based on the gemini-2.5-flash model with additional tuning to address question and answering tasks. The model is frozen after release.	128K	June 17, 2026
gemini-2.0-flash-001/answer_gen/v1	A Gemini Enterprise model based on the gemini-2.0-flash-001 model with additional tuning to address question and answering tasks. The model is frozen after release.	128K	Feb 5, 2026
preview	The preview model specification points to the gemini-2.5-flash model. The preview model is subject to change without notification. If you use preview as the model, you might see changes in the responses when the model changes. If you want consistency in the responses, select a specific model.	128K	N/A

What's next

• Get search summaries