Stay organized with collections
Save and categorize content based on your preferences.
Create a custom translation model
Train and use a custom translation model by using the Google Cloud console. The
following example uses AutoML Translation to train an English-to-Spanish
translation model by using a dataset that contains technology-oriented segment
pairs from software localization.
Before you begin
Before you can start using AutoML Translation, your project must have the
Cloud Translation API enabled, and you must have the permissions that are granted by
the following roles:
Viewer role to view existing resources in your project
Cloud Translation API Editor role to create and manage datasets and
models
Storage Admin role to upload training data to a Cloud Storage
bucket
Create a translation dataset and import segment pairs
Download the
archive file that contains the sample data for training the model, and
extract the files.
For this tutorial, you'll use the English to Spanish TSV file.
From the navigation pane, click Datasets to go to the Datasets page.
Click Create dataset.
In the Create dataset dialog, specify details about the dataset:
Enter tutorial_dataset as the name for the dataset.
Select English (EN) as your source language from the drop-down list.
Select Spanish (ES) as your target language.
Click Create.
After the dataset is created, click the dataset name to view its details.
Go to the Import tab and upload the en-es.tsv dataset to
Cloud Storage:
Select Upload files from your computer.
Click Select files, and choose the en-es.tsv file that you
previously downloaded and extracted.
Click Browse to select or create a new Cloud Storage bucket
where your TSV is stored. The bucket region must be us-central1.
Click Continue.
AutoML Translation automatically splits your data into training,
validation, and testing sets. You can view these splits and the imported
sentence pairs in the Sentences tab of your dataset.
In the Previous evaluations section, Cloud Translation shows your model's
BLEU score compared to the Google NMT model. The BLEU (Bilingual Evaluation
Understudy)
score indicates how similar the candidate text is to the reference
texts; values closer to 100 represent more similar texts.
Use the translation model
From the Google Cloud console, you can use your custom model to translate some
text.
In the English text box, enter text to translate and then click
Translate.
You can compare the results from your custom model to the Google NMT model.
Clean up
To avoid unnecessary Google Cloud charges, delete your model,
dataset, and en-es.tsv file. You can also use the
Google Cloud console to delete your project if you don't need it.
[[["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."],[],[],null,["# Create a custom translation model\n=================================\n\n| **Note:** Translation LLM can be customized with your training data using [Vertex AI supervised fine-tuning - Public Preview](/vertex-ai/generative-ai/docs/models/translation-supervised-tuning).\n\nTrain and use a custom translation model by using the Google Cloud console. The\nfollowing example uses AutoML Translation to train an English-to-Spanish\ntranslation model by using a dataset that contains technology-oriented segment\npairs from software localization.\n| **Note:** The following tutorial assumes that, for your project, the Google Cloud console is using the Cloud Translation API instead of the AutoML API to create datasets. This condition is true if you have at least one native Cloud Translation resource or no legacy AutoML resources in your project. If you have only legacy AutoML resources, see [Upgrade AutoML resources](/translate/docs/advanced/automl-upgrade) for more information.\n\nBefore you begin\n----------------\n\nBefore you can start using AutoML Translation, your project must have the\nCloud Translation API enabled, and you must have the permissions that are granted by\nthe following roles:\n\n- **Viewer** role to view existing resources in your project\n- **Cloud Translation API Editor** role to create and manage datasets and models\n- **Storage Admin** role to upload training data to a Cloud Storage bucket\n\nCreate a translation dataset and import segment pairs\n-----------------------------------------------------\n\n1. [Download](/static/translate/docs/advanced/sample/automl-translation-data.zip) the\n archive file that contains the sample data for training the model, and\n extract the files.\n\n For this tutorial, you'll use the English to Spanish TSV file.\n2. Go to the AutoML Translation console.\n\n [Go to the\n Translation page](https://console.cloud.google.com/translation)\n3. From the navigation pane, click **Datasets** to go to the **Datasets** page.\n\n4. Click **Create dataset**.\n\n5. In the **Create dataset** dialog, specify details about the dataset:\n\n 1. Enter `tutorial_dataset` as the name for the dataset.\n 2. Select **English (EN)** as your source language from the drop-down list.\n 3. Select **Spanish (ES)** as your target language.\n 4. Click **Create**.\n6. After the dataset is created, click the dataset name to view its details.\n\n7. Go to the **Import** tab and upload the `en-es.tsv` dataset to\n Cloud Storage:\n\n 1. Select **Upload files from your computer**.\n 2. Click **Select files** , and choose the `en-es.tsv` file that you previously downloaded and extracted.\n 3. Click **Browse** to select or create a new Cloud Storage bucket where your TSV is stored. The bucket region must be `us-central1`.\n8. Click **Continue**.\n\n AutoML Translation automatically splits your data into training,\n validation, and testing sets. You can view these splits and the imported\n sentence pairs in the **Sentences** tab of your dataset.\n\nTrain a model\n-------------\n\n1. Go to the AutoML Translation console.\n\n [Go to the\n Translation page](https://console.cloud.google.com/translation)\n2. From the navigation pane, go to the **Datasets** page.\n\n3. Click the **tutorial_dataset** dataset.\n\n4. Go to the **Train** tab.\n\n5. Click **Start training** , which opens the **Train new model** pane.\n\n6. Enter `tutorial_model` for the model name.\n\n7. Click **Start training**.\n\nTraining a model can take several hours to complete.\n\nEvaluate the model\n------------------\n\nCheck to see how the model compares to the default Google NMT model that is\nbased on segment pairs from your test set.\n\n1. Go to the AutoML Translation console.\n\n [Go to the\n Translation page](https://console.cloud.google.com/translation)\n2. From the navigation pane, go to the **Models** page.\n\n3. Click the **tutorial_model** model.\n\n4. Click the **Evaluate** tab.\n\nIn the **Previous evaluations** section, Cloud Translation shows your model's\nBLEU score compared to the Google NMT model. The [BLEU (Bilingual Evaluation\nUnderstudy)](/translate/docs/advanced/automl-evaluate#bleu)\nscore indicates how similar the candidate text is to the reference\ntexts; values closer to 100 represent more similar texts.\n\nUse the translation model\n-------------------------\n\nFrom the Google Cloud console, you can use your custom model to translate some\ntext.\n\n1. Go to the AutoML Translation console.\n\n [Go to the\n Translation page](https://console.cloud.google.com/translation)\n2. From the navigation pane, go to the **Models** page.\n\n3. Click the **tutorial_model** model.\n\n4. Click the **Predict** tab.\n\n5. In the **English** text box, enter text to translate and then click\n **Translate**.\n\n You can compare the results from your custom model to the Google NMT model.\n\nClean up\n--------\n\nTo avoid unnecessary Google Cloud charges, delete your [model](/translate/docs/advanced/automl-models#delete-model),\n[dataset](/translate/docs/advanced/automl-datasets#delete-dataset), and `en-es.tsv` file. You can also use the\n[Google Cloud console](https://console.cloud.google.com/) to delete your project if you don't need it.\n\nWhat's next\n-----------\n\n- To learn about custom models, see the [Beginner's guide](/translate/docs/advanced/automl-beginner).\n- To create your own dataset and custom model, see [Prepare training\n data](/translate/docs/advanced/automl-prepare) for instructions on how to prepare your data."]]