Fetching paper information

[Benjamin Bialuchukwu Bakwenye]

Hi, 

Can anyone help. I have a list of papers around 115k paperIds, I want to get paper information for each paperId using API. I do not have memory for snapshot as i reckon it would be the best approach. 

I tried:

1. Multiprocessing but this won't help as the API limit is 100k per day and 10 per minute. 
2. I read somewhere that I can send the paperId in batches of 50, and that could work. Tried it but failed. 

Can anyone help please.

Kind regards, 

Benjamin

[Samuel Mok]

Hi Benjamin,

Did you check out the rate limit & auth page on the openalex docs? It has a link to the code example showing how to retrieve 50 ids at a time, and also how you can get in touch to raise the API limit if you're a researcher (or how to pay for premium access otherwise).

Cheers,

Samuel

--
You received this message because you are subscribed to the Google Groups "OpenAlex Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openalex-commun...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/openalex-community/7e8651b8-4e64-4459-acdd-507c98c85fc0n%40googlegroups.com.

[Krugs.de]

Are you using R? OpenalexR does automatic batching when you provide multiple DOIs (used it and it works nicely) and I would assume also for OpenAlex ids. Otherwise you could look at the code and see how they did it?

Rainer

On 16 Apr 2025, at 18:34, Benjamin Bialuchukwu Bakwenye <benjamin...@bse.eu> wrote:

Hi, 

--

[Gabor Schubert]

Hi Benjamin,

The API rate limit is 10 per SECONDS, not 10 per minutes. And to fetch 115k papers in a batch of 50s should be 115000/50=2300 API calls which is far less than the daily limit of 100 000. I did several such large queries without a problem. It can take a second or more to execute an API call to fetch 50 items so there is not a chance that you reach the 10 calls per second limit. 

Best regards,

Gabor Schubert

Stockholm University