KernelHttpServer sample app returns 404

[andreas-deruiter]

Describe the bug KernelHttpServer is calling the Chat endpoint instead of the Text completion endpoint. Because of this, sample apps like the Book Creator App calling it get the following:

Invalid request: The request is not valid, HTTP status: 404 - Detail: This is not a chat model and thus not supported in the v1/chat/completions endpoint. Did you mean to use v1/completions? Status: 404 (Not Found)

Content: { "error": { "message": "This is not a chat model and thus not supported in the v1/chat/completions endpoint. Did you mean to use v1/completions?", "type": "invalid_request_error", "param": "model", "code": null } }

Headers: Date: Thu, 01 Jun 2023 16:56:14 GMT Connection: keep-alive Access-Control-Allow-Origin: REDACTED openai-organization: REDACTED openai-processing-ms: REDACTED openai-version: REDACTED Strict-Transport-Security: REDACTED x-ratelimit-limit-requests: REDACTED x-ratelimit-limit-tokens: REDACTED x-ratelimit-remaining-requests: REDACTED x-ratelimit-remaining-tokens: REDACTED x-ratelimit-reset-requests: REDACTED x-ratelimit-reset-tokens: REDACTED X-Request-ID: REDACTED CF-Cache-Status: REDACTED Server: cloudflare CF-RAY: REDACTED Alt-Svc: REDACTED Content-Type: application/json Content-Length: 236

This bug was probably introduced in PR #1279. SemanticKernelFactory.cs line 49 should call AddOpenAITextCompletionService() instead of AddOpenAIChatCompletionService(). Line 54 should call AddAzureTextCompletionService() instead of AddAzureChatCompletionService()

[craigomatic]

We updated the samples to use chat completion models, if you use gpt35 turbo or gpt4 you should not see this error.

[andreas-deruiter]

OK, got it, thank you!