Stay organized with collections
Save and categorize content based on your preferences.
The Agent Assist Summarization feature lets you provide
conversation summaries to your agents after each conversation is completed.
The summaries help agents create their conversation notes and understand
end-user communication history. For example, a summary output about a
conversation might look similar to the following:
This tutorial guides you through training and deploying a Summarization model
using the Agent Assist console. You can use it to train a model
and test its performance, but be aware that all runtime operations must be
carried out by calling the API directly. See the Agent Assist
Summarization how-to guide for instructions.
If you are using your own data, make sure that you have
formatted it correctly
and uploaded it to a Cloud Storage bucket. You also have the option of
training a model using demo chat data or using a pre-trained demo model.
Create & train a new model
Navigate to the Agent Assist console.
Select the Summarization card in the center of the screen and click
Get started. You have the option of trying out the Summarization feature
using a demo model, or creating your own custom model using one or more
datasets.
If you are training a custom model using the public Summarization dataset,
enter gs://summarization_integration_test_data/data/* in the dataset URI
field. If you are using your own dataset, the tutorial will walk you through
the process of creating a dataset from your data.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-28 UTC."],[[["\u003cp\u003eAgent Assist Summarization provides conversation summaries to agents after each conversation, aiding in note-taking and understanding communication history.\u003c/p\u003e\n"],["\u003cp\u003eYou can train and deploy a Summarization model using the Agent Assist console, but runtime operations require direct API calls.\u003c/p\u003e\n"],["\u003cp\u003eTraining a custom model requires properly formatted data uploaded to a Cloud Storage bucket, but demo data or a pre-trained model is also available.\u003c/p\u003e\n"],["\u003cp\u003eThe Summarization feature can be initiated via the Agent Assist console, where you can either try a demo model or create a custom model using datasets.\u003c/p\u003e\n"],["\u003cp\u003eThis is a pre-GA feature, therefore it is available "as is" with limited support as per the "Pre-GA Offerings Terms".\u003c/p\u003e\n"]]],[],null,["# Train a Summarization custom model for chat\n\n| **Preview**\n|\n|\n| This feature is subject to the \"Pre-GA Offerings Terms\" in the General Service Terms section\n| of the [Service Specific Terms](/terms/service-terms#1).\n|\n| Pre-GA features are available \"as is\" and might have limited support.\n|\n| For more information, see the\n| [launch stage descriptions](/products#product-launch-stages).\n\nThe **Agent Assist Summarization** feature lets you provide\nconversation summaries to your agents after each conversation is completed.\nThe summaries help agents create their conversation notes and understand\nend-user communication history. For example, a summary output about a\nconversation might look similar to the following:\n\nThis tutorial guides you through training and deploying a Summarization model\nusing the Agent Assist console. You can use it to train a model\nand test its performance, but be aware that all runtime operations must be\ncarried out by calling the API directly. See the Agent Assist\n[Summarization how-to guide](/agent-assist/docs/summarization) for instructions.\n\nIf preferred, you can also create and deploy a Summarization model by\n[calling the API directly](/agent-assist/docs/summarization)\n\nBefore you begin\n----------------\n\n1. If you are using your own data, make sure that you have [formatted it correctly](/agent-assist/docs/summarization#summarization_training_data) and uploaded it to a Cloud Storage bucket. You also have the option of training a model using demo chat data or using a pre-trained demo model.\n\nCreate \\& train a new model\n---------------------------\n\nNavigate to the [Agent Assist console](https://agentassist.cloud.google.com).\nSelect the Summarization card in the center of the screen and click\n**Get started**. You have the option of trying out the Summarization feature\nusing a demo model, or creating your own custom model using one or more\ndatasets.\n\nIf you are training a custom model using the public Summarization dataset,\nenter `gs://summarization_integration_test_data/data/*` in the dataset URI\nfield. If you are using your own dataset, the tutorial will walk you through\nthe process of creating a dataset from your data.\n\nWhat's next\n-----------\n\nAfter you have deployed your model, you can then proceed to\n[create a conversation profile](/agent-assist/docs/conversation-profile)."]]
