Plugin not working

[arozehnal]

Obsidian v1.2.7
Copilot v.1.2
Valid API key, newly acquired GPT+

First try - GPT-4
Console output.
Failed to load resource: the server responded with a status of 404 ()
plugin:copilot:37818 Error in streamManager.streamSSE:
CustomEvent
data: "{\n "error": {\n "message": "The model: gpt-4 does not exist",\n "type": "invalid_request_error",\n "param": null,\n "code": "model_not_found"\n }\n}\n"
isTrusted: false
source: SSE2 {INITIALIZING: -1, CONNECTING: 0, OPEN: 1, CLOSED: 2, url: 'https://api.openai.com/v1/chat/completions', …}
bubbles: false
cancelBubble: false
cancelable: false
composed: false
currentTarget: null
defaultPrevented: false
detail: null
eventPhase: 0
path: []
returnValue: true
srcElement: null
target: null
timeStamp: 89358.5
type: "error"
[[Prototype]]: CustomEvent
getAIResponse @ plugin:copilot:37818

Second try - GPT 3.5
Console output
Error in streamManager.streamSSE:
CustomEvent {isTrusted: false, data: '{\n "error": {\n "message": "You exceeded … "param": null,\n "code": null\n }\n}\n', source: SSE2, detail: null, type: 'error', …}
data: "{\n "error": {\n "message": "You exceeded your current quota, please check your plan and billing details.",\n "type": "insufficient_quota",\n "param": null,\n "code": null\n }\n}\n"
isTrusted: false
source: SSE2 {INITIALIZING: -1, CONNECTING: 0, OPEN: 1, CLOSED: 2, url: 'https://api.openai.com/v1/chat/completions', …}
bubbles: false
cancelBubble: false
cancelable: false
composed: false
currentTarget: null
defaultPrevented: false
detail: null
eventPhase: 0
path: []
returnValue: true
srcElement: null
target: null
timeStamp: 29311
type: "error"
[[Prototype]]: CustomEvent
detail: (...)
initCustomEvent: ƒ initCustomEvent()
constructor: ƒ CustomEvent()
Symbol(Symbol.toStringTag): "CustomEvent"
bubbles: (...)
cancelBubble: (...)
cancelable: (...)
composed: (...)
currentTarget: (...)
defaultPrevented: (...)
eventPhase: (...)
path: (...)
returnValue: (...)
srcElement: (...)
target: (...)
timeStamp: (...)
type: (...)
get detail: ƒ detail()
[[Prototype]]: Event

[logancyang]

From your error messages it seems you don't have access to GPT4 yet? The gpt3.5 error seems to read that "you exceeded your current quota". So it's probably the issue with your OpenAI access. Can you check again when you fix that? I may need to add better OpenAI error messages to be more clear.

Make sure you can get a response with a simple "hello" message from gpt3.5.

[arozehnal]

The response with a simple "hello" message from gpt3.5 is OK now.

See an attachment/console log for "hello" message. from gpt4.
gpt4_helllo_console.txt

[logancyang]

Thanks for providing the console messages. Can you please confirm that you have access to GPT-4 by going to https://platform.openai.com/playground?mode=chat&model=gpt-4 and get some response there? Let me know!

Also I've been seeing issues where GPT-4 API is simply down and returning 500. Currently I have to make the OpenAI failure messages better in this plugin to make it easier to troubleshoot.

[logancyang]

Just made a new release 1.2.1 with better error message like this, can you update the plugin and check the error message again? Let me know if more details are needed in troubleshooting OpenAI issues.

[arozehnal]

WWW page https://platform.openai.com/playground?mode=chat&model=gpt-4 reports -> The model: gpt-4 does not exist
I misunderstood about the OpenAI subscription, I probably need to sign up for the whitelist, right?

[logancyang]

WWW page https://platform.openai.com/playground?mode=chat&model=gpt-4 reports -> The model: gpt-4 does not exist I misunderstood about the OpenAI subscription, I probably need to sign up for the whitelist, right?

Yes, you don't automatically get access to the GPT4 API by buying ChatGPT Plus, you still need to sign up for the waitlist afaik.

When you get access to the API, you may unsub from the Plus and only pay for what you use using the API and this plugin. Hope this helps. Closing this issue.