Hello everyone. I’m trying to use the new Deep Research API, but I keep getting an error saying I’ve exceeded the 200,000-token limit.
Error log:
[429] Rate limit reached for o4-mini-deep-research in organization org-XXXXXXXXXX on tokens per min (TPM): Limit 200000, Used 176402, Requested 23675. Please try again in 23ms. Visit https://platform.openai.com/account/rate-limits to learn more.
But it’s strange—no matter how small the request is, the limit is always hit when using Deep Research. How do you work with it?
P.S. image is a screenshot from make using o4-mini-deep-research API
1 Like
Out of interest what tier of customer are you?
But he reasoning, search, consolidation steps won’t be.
Tier-1 qualification. , 5$ pay.
But I didn’t looked for info how rate limit of TPM will be increased after switching to tier-2.
Here is all info about rate limits.
I suspect you will need to be patient and spend some more money and rank up before deep research TPM becomes practical.
It does seem a bit silly that Tier 1 has access to this at all if TPM is preclusive.
There’s limits on every tier…
Honestly, the more you spend and use what you have access to the more you can use it.
Try hitting the limits you have prior to the month being up. Then you will auto move up and or send support a chat from your openAI platform dash and request it.
Do “get around it” now you could use Agents SDK + Tools / MCP and prompt engineering.
The concept of “deep research” is not limited to only their API for it. I’ve found MUCH BETTER results using the OpenAI agents SDK with mix and matching models and tools.
I’m a tier 5 customer and have been for several years.
How is that relevant to the OPs issue?
