Discussions

Ask a Question

Can we download a model after finetuning?

please describe me how can i download a model?

how to send the files to rag using api

url = "<https://api.gradient.ai/api/rag-collections/id/files"> how send the files to api

how do i export a LMM

how do i export a LMM <br>

Maximum requests per minute

Do you have a throughput limit when sending requests using API or SDK?

fine tuning

Can we fine tune llama2 model on single cpu?

how to find the suitable num_epochs in fine-tuning?

I have 750 lines, and I seperate it into 38 sub chunks. I train the all chunks once. Do I need to train the all chunks for couple times by setting the num_epochs in the fine-tuning? if yes, how large is a suitable number?

RAG File Format?

Is there an optimum file format for a single RAG text format? I am currently dumping in a JSON text file and didn't know if there was a better format.

Can we download a model after finetuning?

I can finetune a model and tested model its working as expected, can I download a model that was fintunned?

how to delete a RAG collection

how can i delete a RAG collection if i dont want to use it any more

where to get Bearer token

On this page to test out the Complete Model endpoint it says to input Bearer Where do I get the Bearer? the access tokens do not work <https://docs.gradient.ai/reference/completemodel>