The Rate Limit is ver low! how do you manage an application with these low rate limits? The maximum is 15 img/min and if you have just 100 user in one minute that will not work at all! what’s the best solution?
I’m seeing 50/images a minute here in the updated FAQ page, but this is for DALLE2…
The default rate limit for the DALL·E API is 50 images per minute.
If you dig a bit deeper, you’ll see the rate limits are based on your Tier…which is based on how much you’ve paid and how long you’ve been paying.
You can view the rate and usage limits for your organization under the limits section of your account settings. As your usage of the OpenAI API and your spend on our API goes up, we automatically graduate you to the next usage tier. This usually results in an increase in rate limits across most models. Organizations in higher tiers also get access to lower latency models.
Tier Qualification Usage limits Free User must be in an allowed geography $100 / month Tier 1 $5 paid $100 / month Tier 2 $50 paid and 7+ days since first successful payment $500 / month Tier 3 $100 paid and 7+ days since first successful payment $1,000 / month Tier 4 $250 paid and 14+ days since first successful payment $5,000 / month Tier 5 $1,000 paid and 30+ days since first successful payment $10,000 / month
Tier 5 rate limits
This is a high level summary and there are per-model exceptions to these limits (e.g. some legacy models or models with larger context windows have different rate limits). To view the exact rate limits per model for your account, visit the limits section of your account settings.
Model RPM RPD TPM
10,000 - 300,000
10,000 - 1,000,000
10,000 - 10,000,000
500 - -
500 - -
500 img/min - -
As it shows 15 images a minute for DALLE3 even at Tier 5 now, I imagine it’s so that any one customer doesn’t bring it all crashing down. With that said, as things move forward, I’m sure it will likely increase. If you have a winning app with a lot of traffic, you might also look at enterprise Azure offerings.
Here’s some good rate-limit advice from OpenAI …