A list of GroundingChunk objects. Each chunk represents a piece of retrieved content (for example, from a web page). that the model used to ground its response.
Google Search entry point for web searches. This contains an HTML/CSS snippet that must be embedded in an app to display a Google Search entry point for follow-up web searches related to a model's "Grounded Response".
A list of web search queries that the model performed to gather the grounding information. These can be used to allow users to explore the search results themselves.
GroundingMetadata.groundingChunks
A list of GroundingChunk objects. Each chunk represents a piece of retrieved content (for example, from a web page). that the model used to ground its response.
Signature:
groundingChunks?:GroundingChunk[];
GroundingMetadata.groundingSupports
A list of GroundingSupport objects. Each object details how specific segments of the model's response are supported by the groundingChunks.
Google Search entry point for web searches. This contains an HTML/CSS snippet that must be embedded in an app to display a Google Search entry point for follow-up web searches related to a model's "Grounded Response".
Signature:
searchEntryPoint?:SearchEntrypoint;
GroundingMetadata.webSearchQueries
A list of web search queries that the model performed to gather the grounding information. These can be used to allow users to explore the search results themselves.
[null,null,["Last updated 2025-07-17 UTC."],[],[],null,["Metadata returned when grounding is enabled.\n\nCurrently, only Grounding with Google Search is supported (see [GoogleSearchTool](./ai.googlesearchtool.md#googlesearchtool_interface)).\n| **Important:** If using Grounding with Google Search, you are required to comply with the \"Grounding with Google Search\" usage requirements for your chosen API provider: [Gemini Developer API](https://ai.google.dev/gemini-api/terms#grounding-with-google-search) or Vertex AI Gemini API (see [Service Terms](https://cloud.google.com/terms/service-terms) section within the Service Specific Terms).\n\n**Signature:** \n\n export interface GroundingMetadata \n\nProperties\n\n| Property | Type | Description |\n|-----------------------------------------------------------------------------------|-----------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| [groundingChunks](./ai.groundingmetadata.md#groundingmetadatagroundingchunks) | [GroundingChunk](./ai.groundingchunk.md#groundingchunk_interface)\\[\\] | A list of [GroundingChunk](./ai.groundingchunk.md#groundingchunk_interface) objects. Each chunk represents a piece of retrieved content (for example, from a web page). that the model used to ground its response. |\n| [groundingSupports](./ai.groundingmetadata.md#groundingmetadatagroundingsupports) | [GroundingSupport](./ai.groundingsupport.md#groundingsupport_interface)\\[\\] | A list of [GroundingSupport](./ai.groundingsupport.md#groundingsupport_interface) objects. Each object details how specific segments of the model's response are supported by the `groundingChunks`. |\n| [retrievalQueries](./ai.groundingmetadata.md#groundingmetadataretrievalqueries) | string\\[\\] | |\n| [searchEntryPoint](./ai.groundingmetadata.md#groundingmetadatasearchentrypoint) | [SearchEntrypoint](./ai.searchentrypoint.md#searchentrypoint_interface) | Google Search entry point for web searches. This contains an HTML/CSS snippet that must be embedded in an app to display a Google Search entry point for follow-up web searches related to a model's \"Grounded Response\". |\n| [webSearchQueries](./ai.groundingmetadata.md#groundingmetadatawebsearchqueries) | string\\[\\] | A list of web search queries that the model performed to gather the grounding information. These can be used to allow users to explore the search results themselves. |\n\nGroundingMetadata.groundingChunks\n\nA list of [GroundingChunk](./ai.groundingchunk.md#groundingchunk_interface) objects. Each chunk represents a piece of retrieved content (for example, from a web page). that the model used to ground its response.\n\n**Signature:** \n\n groundingChunks?: GroundingChunk[];\n\nGroundingMetadata.groundingSupports\n\nA list of [GroundingSupport](./ai.groundingsupport.md#groundingsupport_interface) objects. Each object details how specific segments of the model's response are supported by the `groundingChunks`.\n\n**Signature:** \n\n groundingSupports?: GroundingSupport[];\n\nGroundingMetadata.retrievalQueries\n\u003e | **Warning:** This API is now obsolete.\n\u003e\n\u003e Use [GroundingSupport](./ai.groundingsupport.md#groundingsupport_interface) instead.\n\n**Signature:** \n\n retrievalQueries?: string[];\n\nGroundingMetadata.searchEntryPoint\n\nGoogle Search entry point for web searches. This contains an HTML/CSS snippet that must be embedded in an app to display a Google Search entry point for follow-up web searches related to a model's \"Grounded Response\".\n\n**Signature:** \n\n searchEntryPoint?: SearchEntrypoint;\n\nGroundingMetadata.webSearchQueries\n\nA list of web search queries that the model performed to gather the grounding information. These can be used to allow users to explore the search results themselves.\n\n**Signature:** \n\n webSearchQueries?: string[];"]]
