Put your API key into a text file named something like this: “api_key.txt”
Then type openai.api_key_path=“api_key.txt”
You can also use base64 encryption to encrypt the api key, but you don’t need to.
2 Likes
If you run openai using command ! openai, you should set api-key on environtment variable of your OS or you can set using this script:
import os
os.environ[‘OPENAI_API_KEY’] = ‘your key’
2 Likes
Create a new key and wait a little it will generate whole key and you will be able to copy it only once, if you loose u have to generate new one!
arjava
8
that’s work for my error in python3.10
ganga
9
I have tried both using api_key=
os.environ[“OPENAI_API_KEY”] = api_key
and export OPENAI_API_KEY-
also.
Still i’m getting the same error.
Whether API Key should be used same or can be changed to new every time we run the trained model
1 Like
Anser
10
import openai
openai.api_key = “…”
!export OPENAI_API_KEY=“…”
!openai api fine_tunes.create -t /content/drive/My\ Drive/trainingPromptsAndCompletions1.jsonl -m davinci
I have tried both ways, still getting same error:
Error: No API key provided. You can set your API key in code using ‘openai.api_key = ’, or you can set the environment variable OPENAI_API_KEY=). If your API key is stored in a file, you can point the openai module at it with ‘openai.api_key_path = ’. You can generate API keys in the OpenAI web interface. See https://onboard.openai.com for details, or email support@openai.com if you have any questions.
Has anyone solved it yet?
I have never had a single problem using the environmental variable OPENAI_API_KEY, but I don’t use the OpenAI CLI and deleted it off my workstation within minutes after installing it.
HTH

I use PyCharm to write code on Windows 11, and I have the same problem.
The problem disappeared after restarting PyCharm.
No @ganga , i written this, it is working fine
os.environ[‘OPENAI_API_KEY’] = key
!openai api fine_tunes.create -t “sport2_prepared_train.jsonl” -v “sport2_prepared_valid.jsonl” --compute_classification_metrics --classification_positive_class “baseball” -m ada
i think you don’t understand his question.
his prompt is not python script but native prompt linux
then :
$ openai -h
usage: openai [-h] [-v] [-b API_BASE] [-k API_KEY] [-o ORGANIZATION] {api,tools,wandb} …
positional arguments:
{api,tools,wandb}
api Direct API calls
tools Client side tools for convenience
wandb Logging with Weights & Biases
options:
-h, --help show this help message and exit
-v, --verbose Set verbosity.
-b API_BASE, --api-base API_BASE
What API base url to use.
-k API_KEY, --api-key API_KEY
What API key to use.
-o ORGANIZATION, --organization ORGANIZATION
Which organization to run as (will use your default organization if not specified)
ex: “openai -k my_api_key api fine_tunes.create -t /home/dataset_prepared.jsonl -m gpt-3.5-turbo”
1 Like
You should use it directly in code (without os.getenv() )
1 Like
you can solve this issue simply just by replacing your with code given below:
openai.api_key = “your-key”
1 Like
Try this out in your command line:
openai -k <YOUR_API_KEY> api fine_tunes.create -t “your_path_to_gpt_prepared.jsonl” -m
1 Like
Separating into different cell the execution in kaggle solved for me.
that’s work for my error Thank you
Hello,
I’m Anna and I am new in this area. I have the same error as you @mp .
When I try to finetune a model I get thes message: >
openai api fine_tunes.follow -i ft-xxxxxx
←[91mError:←[0m No API key provided. You can set your API key in code using ‘openai.api_key = ’, or you can set the environment variable OPENAI_API_KEY=). If your API key is stored in a file, you can point the openai module at it with ‘openai.api_key_path = ’. You can generate API keys in the OpenAI web interface. See the onbard page from openai for details, or email support@openai.com if you have any questions.
How did you solve the problem ?
Unfortunately I could not find it in your answers…
I would be very grateful for help
pg3001
25
In my case I basically had to move the
import os
os.environ["OPENAI_API_KEY"] = <OPENAI_API_KEY>
to the very top of the file ABOVE OTHER IMPORTS and it basically works.
2 Likes
That solved it for me - the import of openai itself had to come after the environment setting:
import os
os.environ['OPENAI_API_KEY'] = 'your key here'
import openai
blah blah blah...