Total Cached Content Storage Tokens Per Model Free Tier limit exceeded

Gemini API
,
1

Hi, everyone. I got below error, and I am confused. I am using a tier 1 api key instead of free tier. and I cant see there is such quota in api cloud. I tried to list or delete existing caches, but turns out the list return nothing. Anyone knows how to deal with it? Thanks

Failed Precondition: 429 RESOURCE_EXHAUSTED. {‘error’: {‘code’: 429, ‘message’: ‘TotalCachedContentStorageTokensPerModelFreeTier limit exceeded for model gemini-2.5-flash: limit=0, requested=10158’, ‘status’: ‘RESOURCE_EXHAUSTED’}}

2

The cache quota measured in tokens isn’t currently user-visible. So what we could do in such a situation:-

  1. Switch to models that don’t use tool-calling or system instructions, or reduce memory/tool context wherever possible.
  2. For Gemini2.5 pro, avoid heavy system instructions on the payload if you don’t explicitly need it.
4

@Lai_Steven,

could you please confirm the following,

1.Are you using the same key (Tier one) as that of the key you used while creating it ?
2.can you be sure that the TTL or expiration time is not passed on the caches you mentioned?

7

thanks for reply. unfortunately I do need heavy system instruction. it used to work yesterday, but not now.

9
  1. yes, I only have one key.
  2. yes I didnt.