[Question]: llama-index with gpt4free

[Yusufibin]

Question Validation

• I have searched both the documentation and discord for an answer.

Question

How do I link llama-index to gpt4free?

[Luke-in-the-sky]

I have not seen explicit support for gpt4free, but there seem to be instructions here on how to use the class InstructorEmbeddings to use whichever model you want to

https://gpt-index.readthedocs.io/en/latest/core_modules/model_modules/embeddings/usage_pattern.html#custom-embedding-model

hope that's helpful

[dosubot[bot]]

🤖

Hello,

To link LlamaIndex with gpt4free, you can use the InstructorEmbedding class in the LlamaIndex codebase. This class is used to generate embeddings for text and queries using a specific model, which you can define.

Here is an example of how you might use it:

from llama_index.embeddings import InstructorEmbedding

instructor_embedding = InstructorEmbedding(model_name='gpt4free')
query_embedding = instructor_embedding._get_query_embedding('my query')
text_embedding = instructor_embedding._get_text_embedding('my text')

In this example, an instance of InstructorEmbedding is created with the 'gpt4free' model. Then, embeddings are generated for a query and a text.

Please note that you'll need to replace 'gpt4free' with the actual model name or path of your gpt4free model.

You can find more information about the InstructorEmbedding class in the LlamaIndex codebase.

If you have any other questions or need further clarification, feel free to ask.

Best regards.

Sources

• [Question]: llama-index with gpt4free
• llama_index/embeddings/instructor.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[brycepg]

I got g4f working with llamaindex for queries (not generating embeddings):

https://gist.github.com/brycepg/bab4fd81ae1c93660754746c8ffb4d01

Save as llamag4f.py

Then in your code import it and set it as your llm

import llamag4f
Settings.llm = llamag4f.OurLLM()

Configure g4f model with

Settings.llm = OurLLM(g4f_model="gpt-4")