Starting April 29, 2025, Gemini 1.5 Pro and Gemini 1.5 Flash models are not available in projects that have no prior usage of these models, including new projects. For details, seeModel versions and lifecycle.
Stay organized with collectionsSave and categorize content based on your preferences.
This page provides prerequisites and detailed instructions for fine-tuning
Gemini on text data using supervised learning. For text tuning examples
of classification, sentiment analysis, and extraction use cases,
seeModel tuning for Gemini text models.
Use cases
Fine-tuning lets you adapt base Gemini models for specialized tasks.
Here are some text use cases:
Extracting structured information from chats: Transform multi-turn conversations into organized data by fine-tuning a model to identify key attributes and output them in a structured format like JSONL.
Document categorization: Fine-tune a model to accurately classify lengthy documents into predefined categories, enabling efficient organization and retrieval of information.
Instruction following: Enhance a model's ability to comprehend and execute instructions, leading to more accurate and reliable task completion.
Automated code review: Use fine-tuning to create a model capable of providing insightful code reviews, identifying potential issues, and suggesting improvements.
Summarization: Generate concise and informative summaries of long texts by fine-tuning a model to capture the essence of the content.
Code and DSL generation: Fine-tune a model to generate code in various programming languages or domain-specific languages (DSLs), automating repetitive coding tasks.
Improved RAG performance: Enhance the helpfulness and accuracy of Retrieval-Augmented Generation (RAG) systems by fine-tuning the underlying language model.
Dataset format
ThefileUrifor your dataset can be the URI for a file in a Cloud Storage
bucket, or it can be a publicly available HTTP or HTTPS URL.
{"systemInstruction":{"role":"system","parts":[{"text":"You are a pirate dog named Captain Barktholomew."}]},"contents":[{"role":"user","parts":[{"text":"Hi"}]},{"role":"model","parts":[{"text":"Argh! What brings ye to my ship?"}]},{"role":"user","parts":[{"text":"What's your name?"}]},{"role":"model","parts":[{"text":"I be Captain Barktholomew, the most feared pirate dog of the seven seas."}]}]}
Sample datasets
You can use the following sample datasets to learn how to tune a
Gemini model. To use these datasets, specify the URIs in the
applicable parameters when creating a text model supervised fine-tuning job.
To use the sample tuning dataset, specify its location as follows:
[[["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 2025-09-04 UTC."],[],[],null,["# Text tuning\n\nThis page provides prerequisites and detailed instructions for fine-tuning\nGemini on text data using supervised learning. For text tuning examples\nof classification, sentiment analysis, and extraction use cases,\nsee [Model tuning for Gemini text models](/vertex-ai/generative-ai/docs/models/tune_gemini/tune-gemini-learn).\n\nUse cases\n---------\n\nFine-tuning lets you adapt base Gemini models for specialized tasks.\nHere are some text use cases:\n\n- **Extracting structured information from chats**: Transform multi-turn conversations into organized data by fine-tuning a model to identify key attributes and output them in a structured format like JSONL.\n- **Document categorization**: Fine-tune a model to accurately classify lengthy documents into predefined categories, enabling efficient organization and retrieval of information.\n- **Instruction following**: Enhance a model's ability to comprehend and execute instructions, leading to more accurate and reliable task completion.\n- **Automated code review**: Use fine-tuning to create a model capable of providing insightful code reviews, identifying potential issues, and suggesting improvements.\n- **Summarization**: Generate concise and informative summaries of long texts by fine-tuning a model to capture the essence of the content.\n- **Code and DSL generation**: Fine-tune a model to generate code in various programming languages or domain-specific languages (DSLs), automating repetitive coding tasks.\n- **Improved RAG performance**: Enhance the helpfulness and accuracy of Retrieval-Augmented Generation (RAG) systems by fine-tuning the underlying language model.\n\nDataset format\n--------------\n\nThe `fileUri` for your dataset can be the URI for a file in a Cloud Storage\nbucket, or it can be a publicly available HTTP or HTTPS URL.\n\nThe following is an example of a text dataset.\n\nTo see the generic format example, see\n[Dataset example for Gemini](/vertex-ai/generative-ai/docs/models/gemini-supervised-tuning-prepare#dataset-example). \n\n {\n \"systemInstruction\": {\n \"role\": \"system\",\n \"parts\": [\n {\n \"text\": \"You are a pirate dog named Captain Barktholomew.\"\n }\n ]\n },\n \"contents\": [\n {\n \"role\": \"user\",\n \"parts\": [\n {\n \"text\": \"Hi\"\n }\n ]\n },\n {\n \"role\": \"model\",\n \"parts\": [\n {\n \"text\": \"Argh! What brings ye to my ship?\"\n }\n ]\n },\n {\n \"role\": \"user\",\n \"parts\": [\n {\n \"text\": \"What's your name?\"\n }\n ]\n },\n {\n \"role\": \"model\",\n \"parts\": [\n {\n \"text\": \"I be Captain Barktholomew, the most feared pirate dog of the seven seas.\"\n }\n ]\n }\n ]\n }\n\n### Sample datasets\n\nYou can use the following sample datasets to learn how to tune a\nGemini model. To use these datasets, specify the URIs in the\napplicable parameters when creating a text model supervised fine-tuning job.\n\nTo use the sample tuning dataset, specify its location as follows: \n\n \"training_dataset_uri\": \"gs://cloud-samples-data/ai-platform/generative_ai/gemini-2_0/text/sft_train_data.jsonl\",\n\nTo use the sample validation dataset, specify its location as follows: \n\n \"validation_dataset_uri\": \"gs://cloud-samples-data/ai-platform/generative_ai/gemini-2_0/text/sft_validation_data.jsonl\",\n\nWhat's next\n-----------\n\n- To start tuning, see [Tune Gemini models by using supervised fine-tuning](/vertex-ai/generative-ai/docs/models/gemini-use-supervised-tuning).\n- To learn how supervised fine-tuning can be used in a solution that builds a generative AI knowledge base, see [Jump Start Solution: Generative AI knowledge base](/architecture/ai-ml/generative-ai-knowledge-base)."]]
