Update models to gpt-5 by gewarren · Pull Request #52139 · dotnet/docs

gewarren · 2026-03-11T00:07:37Z

Update to latest models
Model name doesn't need to be a secret

Internal previews

📄 File	🔗 Preview link
docs/ai/conceptual/data-ingestion.md	Data ingestion
docs/ai/evaluation/evaluate-ai-response.md	Quickstart: Evaluate response quality
docs/ai/evaluation/evaluate-safety.md	Tutorial: Evaluate response safety with caching and reporting
docs/ai/evaluation/evaluate-with-reporting.md	Tutorial - Evaluate response quality with caching and reporting
docs/ai/quickstarts/structured-output.md	Request a response with structured output
docs/ai/quickstarts/text-to-image.md	docs/ai/quickstarts/text-to-image

Copilot

Pull request overview

Updates the AI documentation and code snippets to use newer model names (primarily gpt-5) and removes the need to store model names in user-secrets.

Changes:

Updated multiple docs and snippets from gpt-4o to gpt-5.
Removed AZURE_OPENAI_GPT_NAME user-secrets instructions and switched several Azure OpenAI samples to hard-coded deployment names.
Updated tokenizer examples and a few OpenAI (non-Azure) defaults to gpt-5.

Reviewed changes

Copilot reviewed 17 out of 17 changed files in this pull request and generated 13 comments.

Show a summary per file

File	Description
docs/ai/quickstarts/text-to-image.md	Removes the model-name user-secret instruction for the text-to-image quickstart.
docs/ai/quickstarts/structured-output.md	Updates the structured-output quickstart to reference `gpt-5` and removes the model-name user-secret instruction.
docs/ai/quickstarts/snippets/text-to-image/azure-openai/Program.cs	Hard-codes the image deployment name as `gpt-image-1`.
docs/ai/quickstarts/snippets/structured-output/Program.cs	Hard-codes the chat deployment name as `gpt-5`.
docs/ai/quickstarts/snippets/function-calling/openai/Program.cs	Updates the default OpenAI model fallback to `gpt-5`.
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/TiktokenExample.cs	Updates tokenizer model selection to `gpt-5`.
docs/ai/how-to/snippets/handle-invalid-tool-input/csharp/StrictSchema.cs	Updates OpenAI model fallback and generalizes the compatibility comment.
docs/ai/how-to/snippets/handle-invalid-tool-input/csharp/IncludeDetailedErrors.cs	Updates OpenAI model fallback to `gpt-5`.
docs/ai/how-to/snippets/handle-invalid-tool-input/csharp/FunctionInvoker.cs	Updates OpenAI model fallback to `gpt-5`.
docs/ai/how-to/snippets/access-data/ArgumentsExample.cs	Hard-codes the Azure OpenAI deployment name as `gpt-5`.
docs/ai/evaluation/snippets/evaluate-with-reporting/MyTests.cs	Hard-codes the Azure OpenAI deployment name as `gpt-5`.
docs/ai/evaluation/snippets/evaluate-safety/MyTests.cs	Hard-codes the Azure OpenAI deployment name as `gpt-5`.
docs/ai/evaluation/snippets/evaluate-ai-responses/MyTests.cs	Hard-codes the Azure OpenAI deployment name as `gpt-5`.
docs/ai/evaluation/evaluate-with-reporting.md	Updates tutorial text to reference `gpt-5` and removes the model-name user-secret instruction.
docs/ai/evaluation/evaluate-safety.md	Updates tutorial text to reference `gpt-5` and removes the model-name user-secret instruction.
docs/ai/evaluation/evaluate-ai-response.md	Updates quickstart text to reference `gpt-5` and removes the model-name user-secret instruction.
docs/ai/conceptual/data-ingestion.md	Updates tokenizer model selection example to `gpt-5`.

Comments suppressed due to low confidence (2)

docs/ai/quickstarts/snippets/text-to-image/azure-openai/Program.cs:17

model is hard-coded to "gpt-image-1" but is used as the Azure OpenAI deployment name (via GetImageClient). This will break if the deployment name differs. Consider making the deployment name configurable and/or renaming the variable to deploymentName to avoid implying it’s a model ID.

string model = "gpt-image-1";

// Create the Azure OpenAI client and convert to IImageGenerator.
AzureOpenAIClient azureClient = new(
    new Uri(endpoint),

docs/ai/how-to/snippets/access-data/ArgumentsExample.cs:45

model is hard-coded to "gpt-5" but is passed to AzureOpenAIClient.GetChatClient(...), which expects an Azure OpenAI deployment name. This will fail if the deployment name differs. Consider making the deployment name configurable (even if it isn’t secret) and/or renaming the variable to deploymentName.

        string model = "gpt-5";

        // <SnippetUseAdditionalProperties>
        FunctionInvokingChatClient client = new FunctionInvokingChatClient(
            new AzureOpenAIClient(new Uri(endpoint), new AzureKeyCredential(apiKey))

You can also share your feedback on Copilot code review. Take the survey.

Copilot · 2026-03-11T00:13:24Z

docs/ai/quickstarts/structured-output.md

+To provision an Azure OpenAI service and model using the Azure portal, complete the steps in the [Create and deploy an Azure OpenAI Service resource](/azure/ai-services/openai/how-to/create-resource?pivots=web-portal) article. In the "Deploy a model" step, select the `gpt-5` model.

 ## Create the chat app

-Complete the following steps to create a console app that connects to the `gpt-4o` AI model.
+Complete the following steps to create a console app that connects to the `gpt-5` AI model.


This section tells readers to deploy/select the gpt-5 model, but the code snippet calls GetChatClient(deploymentName: ...) with a hard-coded value. Azure OpenAI uses a deployment name that can differ from the model name, so either instruct readers to set the deployment name to gpt-5 or keep the deployment name configurable in the sample.

Copilot · 2026-03-11T00:13:24Z

docs/ai/quickstarts/snippets/structured-output/Program.cs

 string endpoint = config["AZURE_OPENAI_ENDPOINT"];
-string model = config["AZURE_OPENAI_GPT_NAME"];
 string tenantId = config["AZURE_TENANT_ID"];
+string model = "gpt-5";



model is hard-coded to "gpt-5" but is passed as deploymentName. This will fail if the user’s Azure OpenAI deployment name isn’t exactly "gpt-5". Consider reading the deployment name from configuration (even if it isn’t secret) and/or renaming the variable to deploymentName to match how it’s used.

Copilot · 2026-03-11T00:13:25Z

docs/ai/evaluation/evaluate-with-reporting.md

+To provision an Azure OpenAI service and model using the Azure portal, complete the steps in the [Create and deploy an Azure OpenAI Service resource](/azure/ai-services/openai/how-to/create-resource?pivots=web-portal) article. In the "Deploy a model" step, select the `gpt-5` model.

 ## Create the test app

-Complete the following steps to create an MSTest project that connects to the `gpt-4o` AI model.
+Complete the following steps to create an MSTest project that connects to the `gpt-5` AI model.


This section says to select the gpt-5 model, but the accompanying snippet hard-codes the Azure OpenAI deployment name to "gpt-5". Azure deployments can be named differently, so either document that the deployment name must be "gpt-5" or keep it configurable in the sample/tests.

Copilot · 2026-03-11T00:13:25Z

docs/ai/evaluation/evaluate-with-reporting.md

 1. Run the following commands to add [app secrets](/aspnet/core/security/app-secrets) for your Azure OpenAI endpoint, model name, and tenant ID:

   ```bash
   dotnet user-secrets init
   dotnet user-secrets set AZURE_OPENAI_ENDPOINT <your-Azure-OpenAI-endpoint>


The instructions say to add app secrets for the endpoint, model name, and tenant ID, but the commands no longer set any model/deployment name. Add a deployment-name setting (even if it isn’t secret) or update the wording so readers aren’t missing required configuration.

Copilot · 2026-03-11T00:13:25Z

docs/ai/evaluation/evaluate-safety.md

 1. Run the following commands to add [app secrets](/aspnet/core/security/app-secrets) for your Azure OpenAI endpoint, model name, and tenant ID:

   ```bash
   dotnet user-secrets init
   dotnet user-secrets set AZURE_OPENAI_ENDPOINT <your-Azure-OpenAI-endpoint>


The instructions say to add app secrets for the endpoint, model name, and tenant ID, but the commands no longer set any model/deployment name. Add a deployment-name setting (even if it isn’t secret) or update the wording to match the code.

Copilot · 2026-03-11T00:13:26Z

docs/ai/evaluation/snippets/evaluate-with-reporting/MyTests.cs

        string endpoint = config["AZURE_OPENAI_ENDPOINT"];
-        string model = config["AZURE_OPENAI_GPT_NAME"];
        string tenantId = config["AZURE_TENANT_ID"];
+        string model = "gpt-5";

        // Get an instance of Microsoft.Extensions.AI's <see cref="IChatClient"/>


model is hard-coded to "gpt-5" but is passed as the Azure OpenAI deploymentName later in the method. This makes the sample fail unless the deployment is named exactly "gpt-5". Consider reading the deployment name from configuration and/or renaming the variable to deploymentName to reflect how it’s used.

Copilot · 2026-03-11T00:13:26Z

docs/ai/quickstarts/text-to-image.md

 1. Run the following commands to add [app secrets](/aspnet/core/security/app-secrets) for your Azure OpenAI endpoint, model name, and API key:

    ```bash
    dotnet user-secrets init
    dotnet user-secrets set AZURE_OPENAI_ENDPOINT <your-Azure-OpenAI-endpoint>


The instructions say to add app secrets for the endpoint, model name, and API key, but the commands no longer set any model/deployment name. Update the text and/or add a config value for the image deployment name so the setup steps match the Program.cs snippet.

Copilot · 2026-03-11T00:13:26Z

docs/ai/evaluation/evaluate-ai-response.md

+To provision an Azure OpenAI service and model using the Azure portal, complete the steps in the [Create and deploy an Azure OpenAI Service resource](/azure/ai-services/openai/how-to/create-resource?pivots=web-portal) article. In the "Deploy a model" step, select the `gpt-5` model.

 ## Create the test app

-Complete the following steps to create an MSTest project that connects to the `gpt-4o` AI model.
+Complete the following steps to create an MSTest project that connects to the `gpt-5` AI model.


This section tells readers to select the gpt-5 model, but the snippets use a fixed Azure OpenAI deployment name. Azure deployments can be named differently, so either specify that the deployment name must be "gpt-5" or keep the deployment name configurable in the sample.

Copilot · 2026-03-11T00:13:26Z

docs/ai/evaluation/snippets/evaluate-safety/MyTests.cs

        string endpoint = config["AZURE_OPENAI_ENDPOINT"];
-        string model = config["AZURE_OPENAI_GPT_NAME"];
        string tenantId = config["AZURE_TENANT_ID"];
+        string model = "gpt-5";

        // Get an instance of Microsoft.Extensions.AI's <see cref="IChatClient"/>


model is hard-coded to "gpt-5" but is used as the Azure OpenAI deploymentName later in the method. This will break if the deployment name differs. Make the deployment name configurable and/or rename the variable to deploymentName so it’s clear what value is required.

Copilot · 2026-03-11T00:13:27Z

docs/ai/quickstarts/structured-output.md

 1. Run the following commands to add [app secrets](/aspnet/core/security/app-secrets) for your Azure OpenAI endpoint, model name, and tenant ID:

    ```bash
    dotnet user-secrets init
    dotnet user-secrets set AZURE_OPENAI_ENDPOINT <your-Azure-OpenAI-endpoint>


The instructions say to add app secrets for the endpoint, model name, and tenant ID, but the commands no longer set any model/deployment name. Either add a configuration value for the deployment name (even if it isn’t secret) or update the text to remove the model-name requirement so the setup steps match the snippet.

gewarren added 4 commits March 10, 2026 15:52

update models to gpt-5

23a3a7b

pull upstream main

b223987

revert unwanted changes

4b665cf

delete model secret

75d64c1

gewarren requested a review from a team as a code owner March 11, 2026 00:07

Copilot AI review requested due to automatic review settings March 11, 2026 00:07

dotnetrepoman bot added this to the March 2026 milestone Mar 11, 2026

gewarren enabled auto-merge (squash) March 11, 2026 00:08

Copilot started reviewing on behalf of gewarren March 11, 2026 00:08 View session

Copilot AI reviewed Mar 11, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update models to gpt-5#52139

Update models to gpt-5#52139
gewarren wants to merge 4 commits intodotnet:mainfrom
gewarren:gpt-nae

gewarren commented Mar 11, 2026 •

edited by github-actions bot

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Copilot AI Mar 11, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

gewarren commented Mar 11, 2026 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Internal previews

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 11, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

gewarren commented Mar 11, 2026 •

edited by github-actions bot

Loading