Vision restrictions for API - 100 RPD

I want to create a simple private website, which will be able to upload an image, and then use gpt-4-vision API to do further tasks (description, answering questions like “how many people are in the image”) for research reasons first, but if everything goes OK, I want to use this tool for real-life purposes.

My concern is that 100RPD , it means that I won’t be able to send more than 100 requests per day? I will need to use like 1500-1800 requests on a daily basis to be useful for me.

How to get more RPD , or even unlimited RPD ? It’s crucial for me not being limited. The main purpose is to do statistics, how many people are in my shop during given period of the day.
Let’s imagine it as a bakery - I want to be able to do some statistics, based on my built in camera, which sends 1 image per minute, I want to send it to gpt-4-vision and get an integer as a result, to do some graphs, analytics. My employees are complaining, that there are some “peaks” during the day, I am not at the location , also unable to non stop watch the built-in camera in the bakery. Long story short: I want to automate it :slight_smile:

It seems we are all stuck with 100 RPD for now. I checked all tiers and everyone is the same.

From the docs:

  • The models gpt-4-1106-preview and gpt-4-vision-preview are currently under preview with restrictive rate limits that make them suitable for testing and evaluations, but not for production usage. We plan to increase these limits gradually in the coming weeks with an intention to match current gpt-4 rate limits once the models graduate from preview. As these models are adopted for production workloads we expect latency to increase modestly compared to this preview phase.

Reference: Rate Limits

Yep I was reading the same, not so good, let’s hope it will increase soon… I was thinking there is something how to override this. I can create like 20 accounts, fund all of them and rotate the API keys, but it’s a lot of work and not sure if it’s allowed.

this is the error when reaching 100 RPD

API Error: Rate limit reached for gpt-4-vision-preview in organization org-XXXX on requests per day (RPD): Limit 100, Used 100, Requested 1. Please try again in 14m24s. Visit to learn more.

The good news is, that after reachig 100 daily API calls, it cools down for another one each 744 seconds.

Finally, they increased the limits, 80RPM , 500 RPD… I am waiting to be like 10k RPD and it will be widely usable :slight_smile:

Up to 1500 RPD now!

Weird , I still have 500 RPD, so they are updating it per account…

@fero.hora.1969 They should be updating by tier.


Yep, I tiered up to tier 2, it was 100 RPM / 1000 RPD,
then tried tier 3, it is 120 RPM / 1500 RPD.

their tier documentation is still showing wrong numbers, I would be happy to see the higher tier rate limits correctly in that table.


Tier four is up to 2,000 RPD

Hang tight! We are working to increase limits across many of our models, the docs should be updated later today with an up to date look at this by tier, they are slightly out of date right now given we have been pushing updates quickly.

You can follow along in the docs:

And to confirm, rate limits depend what usage tier you are in.


Where are you folks getting info for vision RPD? Perhaps my ole eyes are playing tricks on me and simply overlooking in dashboard?

You’ll need to expand the list with the button “show all models”.


1 Like

Perfect, thank you @_j

Totally missed that.