Hi
We developed a solution where our clients can generate 30 Blog Posts / Social Media Posts at once with text and pictures. while the LLM are no problem within parallelize them, Delle 3 limits are.
In our case, if 2 customers start the process at the sime time, 60 pictures should be generated.
Well, so 50 pics per min are way to low. So my thought are going to use several api keys. But is that the best option? Or do you guys have another solution ?
It would not be two API keys, it would be two accounts and identities, the second also having paid $1000+ and waiting a month (then making another minimum payment for a recalculation).
The better solution is to scroll down to “request an exception” https://platform.openai.com/account/limits - but you find out DALL-E 3 is “other models”, and “We are not currently accepting requests for other models”
(ChatGPT competition not welcome? Microsoft contract exclusivity?)
1 Like
What about Enterprise, would this help in our case, or do enterprise customers have same limits?
1 Like
There is no official “enterprise” for API. You can contact sales and say “we’ve got $100,000 to spend as your new partner”.
Then see if they don’t say, “competing with our $1B product?” 
1 Like
Yeah, multiple accounts gets messy quick.
I’d look into Azure maybe? Not sure if they have DALLE3 yet, though. They do scale a bit better, though?
Currently, Dall-E-3 is only available in the Sweden Central region on Azure.
However, by default, you can only make 2 requests per minute with a single Azure account.
You can increase this limit by providing a detailed explanation of your intended use, but it’s not as straightforward as OpenAI’s usage-based pricing system.
If anyone has successfully requested an increase in usage limits, they might be able to provide some information.
2 Likes