404 error about gpt4-pdf-chatbot-langchain HOT 9 CLOSED

I tried some googling and changed the model to:

modelName: 'gpt-3.5-turbo'

and now it seems to work. @sdugoten - can you try to change the modelName and see if it works for you?

Reference to the docs for a list of supported modelNames

ok, it does seems to be the model name, you have to use "gpt-3.5-turbo" instead of "gpt-3.5". I am gonna try my own pdf and see if that works.

from gpt4-pdf-chatbot-langchain.

For future readers: Yes. You are running into problems with text-davinci-003 because the OpenAI wrapper is hitting the https://api.openai.com/v1/chat/completions endpoint.

These are the supported models for this endpoint:

gpt-4, gpt-4-0314, gpt-4-32k, gpt-4-32k-0314, gpt-3.5-turbo, gpt-3.5-turbo-0301

Source

from gpt4-pdf-chatbot-langchain.

Can you provide more information on the source code behind this error? It appears to be an issue with your API keys.

from gpt4-pdf-chatbot-langchain.

Can you provide more information on the source code behind this error? It appears to be an issue with your API keys.

where can I give you more information so that I can help you to troubleshoot? I just follow the instruction from your main page https://github.com/mayooear/gpt4-pdf-chatbot-langchain and run the program.

I basically generate the openAI key in user setting page, and paste it inside .env

The entry looks something like this

OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxohrzgxxxxx6BM

from gpt4-pdf-chatbot-langchain.

Hey. Thanks for the repo and the work you put in! I have the same error.

  const docChain = loadQAChain(
    new OpenAIChat({
      temperature: 0,
      modelName: 'text-davinci-003', // <--
      streaming: Boolean(onTokenStream),
      callbackManager: onTokenStream
        ? CallbackManager.fromHandlers({
            async handleLLMNewToken(token) {
              onTokenStream(token);
              console.log(token);
            },
          })
        : undefined,
    }),
    { prompt: QA_PROMPT },
  );

This is the only part I changed in the code.

This is the output I get from running curl https://api.openai.com/v1/models -H "Authorization: Bearer XXX"

// ...
    {
      "id": "text-davinci-003",
      "object": "model",
      "created": 1669599635,
      "owned_by": "openai-internal",
      "permission": [
        {
          "id": "modelperm",
          "object": "model_permission",
          "created": 1678585020,
          "allow_create_engine": false,
          "allow_sampling": true,
          "allow_logprobs": true,
          "allow_search_indices": false,
          "allow_view": true,
          "allow_fine_tuning": false,
          "organization": "*",
          "group": null,
          "is_blocking": false
        }
      ],
      "root": "text-davinci-003",
      "parent": null
    },
// ...

These are my pinecone settings. I am not familiar with pinecone, so I copied (or thought that I copied) your values:

from gpt4-pdf-chatbot-langchain.

You can see the error in action here

https://www.youtube.com/watch?v=s82PalNVJiM

from gpt4-pdf-chatbot-langchain.

I tried some googling and changed the model to:

modelName: 'gpt-3.5-turbo'

and now it seems to work. @sdugoten - can you try to change the modelName and see if it works for you?

Reference to the docs for a list of supported modelNames

from gpt4-pdf-chatbot-langchain.

@ankri great find here! Appreciate it. Are you running into errors with text-davinci-003 ?

from gpt4-pdf-chatbot-langchain.

For future readers: Yes. You are running into problems with text-davinci-003 because the OpenAI wrapper is hitting the https://api.openai.com/v1/chat/completions endpoint.

These are the supported models for this endpoint: gpt-4, gpt-4-0314, gpt-4-32k, gpt-4-32k-0314, gpt-3.5-turbo, gpt-3.5-turbo-0301

Source

Good catch!

from gpt4-pdf-chatbot-langchain.