Gemma 3n Input Token Count Limit

Gemma
,
1

When using gemma-3n-e2b-it through the gemini api, I got this error:

‘code’: 400, ‘message’: ‘The input token count (8905) exceeds the maximum number of tokens allowed (8192).’, ‘status’: ‘INVALID_ARGUMENT’

However, both gemma 3n models have context windows of 32k tokens and I haven’t been able to find any documentation indicating the free tier is subject to lower limits. In fact, both Gemma 3n models don’t even appear on the Google AI Studio Rate Limit page in the usage and billing section. Can anyone clarify if there are special limits for these models?

2

Hi @3n_developer, Thank you for bringing this to our attention.

We have reproduced the 400 error using gemma-3n-e2b-it model and we have informed the relevant team.