tabAutocompleteModel After setting it to openai, a message is displayed indicating 404 Not Found

[0x2F4D2]

Before submitting your bug report

• [X] I believe this is a bug. I'll try to join the Continue Discord for questions
• [X] I'm not able to find an open issue that reports the same bug
• [X] I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS: MacOS 14.5
- Continue: v0.8.43
- IDE: Visual Studio Code v1.9.11
- Model: gpt-4o-mini
- config.json:
  
{
  "models": [
    {
      "title": "OpenAI",
      "provider": "openai",
      "model": "gpt-4o-mini",
      "apiKey": "xxx",
      "apiBase": "https://api.openai.com/v1"
    }
  ],
  "tabAutocompleteModel": {
    "title": "OpenAI",
    "provider": "openai",
    "model": "gpt-4o-mini",
    "apiKey": "xxx",
    "apiBase": "https://api.openai.com/v1/completions"
  },
  "useLegacyCompletionsEndpoint": false,
  "allowAnonymousTelemetry": false,
  "embeddingsProvider": {
    "provider": "transformers.js"
  }
}

Description

Chat can be used normally, but CMD + I and code autocomplete, will prompt 404 Error streaming diff: Error: HTTP 404 Not Found from https://api.openai.com/v1/completions You may need to add pre-paid credits before using the OpenAI API.

Actually, I have pre-paid credits.

To reproduce

No response

Log output

No response

[SpeedOfSpin]

I think this only works in the pre-release version currently.

[beb3k]

I've encountered this error too. Only for OpenAI models apparently. If you switch the model (i.e. gemini for example) before clicking apply, it works.

[ImLunaHey]

with the pre-release extension i still get the same 404 error if i use AUTODETECT.

  "tabAutocompleteModel": {
    "model": "AUTODETECT",
    "title": "OpenAI",
    "apiKey": "sk-proj-[REDACTED]",
    "provider": "openai"
  },

if im using the pre-release and use gpt-4o-mini for the model it does work though.

so looks like AUTODETECT is still broken.

[sestinj]

I tested with both of the following in my "models" array in config.json:

{
      "title": "OpenAI",
      "model": "AUTODETECT",
      "provider": "openai",
      "apiKey": "sk-xxx",
      "apiBase": "https://api.openai.com/v1"
    },

and

{
      "title": "OpenAI",
      "model": "gpt-4o-mini",
      "provider": "openai",
      "apiKey": "sk-xxx",
      "apiBase": "https://api.openai.com/v1"
    },

and both worked for CMD + I. This was on both the latest release and pre-release. If anyone is still having the issue with CMD+I I migth need more information to reproduce.

@0x2F4D2 for the autocomplete configuration that you shared there are few things to note:

• useLegacyCompletionsEndpoint belongs inside of the "tabAutocompleteModel" object. You've correctly identified that this is what will force Continue to call the /completions endpoint rather than /chat/completions
• The apiBase should be "https://api.openai.com/v1", not "https://api.openai.com/v1/completions". Continue will add the /completions on its own
• GPT-4 is really bad at autocomplete. If you're willing to use OpenAI, I'd guess you may also be willing to use Mistral. I would highly recommend Codestral, which is much much cheaper and will gives way better results

[landsman]

This error occured to me as well. Config file looks normal.

[github-actions[bot]]

This issue hasn't been updated in 90 days and will be closed after an additional 10 days without activity. If it's still important, please leave a comment and share any new information that would help us address the issue.

[github-actions[bot]]

This issue was closed because it wasn't updated for 10 days after being marked stale. If it's still important, please reopen + comment and we'll gladly take another look!