New local ai quickstart

[alexwolfmsft]

Summary

Describe your changes here.

Fixes #Issue_Number (if available)

---

Internal previews

📄 File	🔗 Preview link
docs/ai/quickstarts/quickstart-local-ai.md	Chat with a local AI model using .NET and Semantic Kernel

[manoj-jain-techporch]

Hi @alexwolfmsft @IEvangelist - Thanks for the document. Very helpful.

But when we tried it, it gives following error in
await foreach (var item in aiChatService.GetStreamingChatMessageContentsAsync(chatHistory))

System.ClientModel.ClientResultException: 'Service request failed. Status: 404 (Not Found)

On the same system, if we try the sample from OlllamaSharp - that works fine on the same model + same port.

If we browse to the http://localhost:11434 it shows "Ollama is running"

Any suggestions?

[alexwolfmsft]

But when we tried it, it gives following error in await foreach (var item in aiChatService.GetStreamingChatMessageContentsAsync(chatHistory))

System.ClientModel.ClientResultException: 'Service request failed. Status: 404 (Not Found)

On the same system, if we try the sample from OlllamaSharp - that works fine on the same model + same port.

If we browse to the http://localhost:11434 it shows "Ollama is running"

Any suggestions?

Thank you for surfacing this - you can find an updated code example here:
https://github.com/dotnet/docs/blob/main/docs/ai/quickstarts/snippets/local-ai/Program.cs

This approach uses the Microsoft.SemanticKernel.Connectors.Ollama package for a better connection experience to Ollama, so you'll need to add that to your project as well.

[manoj-jain-techporch]

Thanks @alexwolfmsft. This worked.