Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

VSCode v0.8.63 - Model URL construction and header "accept" regressions for provider "openai" / apiType "azure" #3427

Open
3 tasks done
Berber31 opened this issue Dec 17, 2024 · 3 comments
Assignees
Labels
area:configuration Relates to configuration options ide:vscode Relates specifically to VS Code extension kind:bug Indicates an unexpected problem or unintended behavior priority:high Indicates high priority

Comments

@Berber31
Copy link
Contributor

Before submitting your bug report

Relevant environment info

- OS:linux
- Continue version: 0.8.63
- IDE version: VSCode
- Model: gpt4o (provider "openai", apiType "azure")
- config.json:
  

  "models" : [ {
    "title" : "GPT-4o-FT",
    "model" : "gpt-4o",
    "provider" : "openai",
    "apiType" : "azure",
    "apiBase" : "https://[my-tenant].openai.azure.com",
    "completionOptions" : {
      "temperature" : 0,
      "topK" : 50,
      "frequencyPenalty" : 0.2
    },
    "requestOptions" : {
      "verifySsl" : false
    }
  } ],

Description

Gpt4o url called is : https://[my-tenant].openai.azure.com/chat/completions
instead of https://[my-tenant].openai.azure.com**/openai/deployments/{engine}/chat/completions?api-version={api-version}**
as expected in the azure documentation https://learn.microsoft.com/en-us/azure/ai-services/openai/reference#rest-api-versioning

Minor:
And 'accept' header was set with '/' in v0.8.61
and now (in v0.8.63) set with 'application/json' (which seems to be not suitable with the azure really returned content-type by azure which is 'text/event-stream')

To reproduce

  1. set confg.json with:
"models" : [ {
    "title" : "GPT-4o-FT",
    "model" : "gpt-4o",
    "provider" : "openai",
    "apiType" : "azure",
    "apiBase" : "https://[my-tenant].openai.azure.com",
    "completionOptions" : {
      "temperature" : 0,
      "topK" : 50,
      "frequencyPenalty" : 0.2
    },
    "requestOptions" : {
      "verifySsl" : false
    }
  } ]
  1. open chat dialog and post a question

Log output

The request failed with "FetchError": request to https://[my-tenant].openai.azure.com/chat/completions failed
@dosubot dosubot bot added area:configuration Relates to configuration options ide:vscode Relates specifically to VS Code extension kind:bug Indicates an unexpected problem or unintended behavior priority:high Indicates high priority labels Dec 17, 2024
@ctwardy
Copy link
Contributor

ctwardy commented Dec 17, 2024

Nice - seems like key info for #3388 - v0.8.63 seems to have re-introduced URL construction errors for Azure OpenAI endpoints.

@Patrick-Erichsen
Copy link
Collaborator

Hi folks, thanks for the notice here. Tracking updates to this issue, and other related issues with Azure after the 0.8.63 release, in this thread: #3477

@sestinj
Copy link
Contributor

sestinj commented Dec 20, 2024

See here: #3477 (comment)

This should be solved in a new release

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area:configuration Relates to configuration options ide:vscode Relates specifically to VS Code extension kind:bug Indicates an unexpected problem or unintended behavior priority:high Indicates high priority
Projects
None yet
Development

No branches or pull requests

5 participants