I’ve been on the waitlist since December… My company is currently using other models like Claude, Perplexity, Poe, finetuneing other models doing very heavy usage per month – we were looking to exclusively use the GPT-4o fine tune
Has anyone had any luck getting access? is there a specifc contact at Open AI to reach out to about our case?
I would contact support at help.openai.com. Due to the nature of the company, you don’t really contact people directly.
Go to the fine-tuning interface in your web account https://platform.openai.com/finetune
Create a new fine tuning by pressing “create”. Select the model gpt-4o.
You will then have “request access” if you have enough fine-tune reputation for consideration.
You’ll likely need to come up with some compelling use case that OpenAI would like to see succeed.
I’ve tried that already and let included our use case — still waiting to hear back… we have in house programers, developers familar with fine-tuning app dev, etc and spend 10k per month using other tools, we would switch exclusively to gpt-4 fine tune
How about fine-tuning GPT-3.5-turbo 5 to 10 times over a few days? It seems that you need to have some experience with fine-tuning to be able to fine-tune GPT-4(o).
tried that as well, but will keep doing it…
If you do not have organization experience fine-tuning, then you do not have the model available.
https://share.hsforms.com/1PsPBTQDRTM6NUZp9J08oCg4sk30
GPT-4 fine-tuning is being developed through an experimental access program. We are eager to understand your specific use cases for GPT-4 fine-tuning to better tailor the product to your needs. Please fill out this form with details of your intended application. We will notify you as soon as the product is ready for wider use.
Filling out the form without a technical plan for developing a training set that could work when overlaid over the GPT-4o chat tuning and therefore succeeding in a novel application would just be wasting everyone’s time.
