recently i started looking into gemma ai models i found them really intressting but im lost i have some questions i really hope you guys can help me
1-can i use the gemma api like gemini in google ai studio ?
2- are there any rate limits
3- what is the pricing of the gemma api
Hi @Mohamed_Amine ,
Please find the answers below for your queries:
Yes, you can access and use the Gemma API through the Gemini API, which is managed by Google AI Studio. This allows you to integrate Gemma models into your applications. Please find the following URL To know more about Gemma with Gemini API.
The Gemma API has rate limits to ensure fair usage and system performance. The specific limits depend on the model, but generally, they are measured in:
A. Requests per minute (RPM)
B. Tokens per minute (TPM)
C. Requests per day (RPD)
For example, for Gemma 3 and 3n models, the rate limits are:
RPM: 30
TPM: 15,000
RPD: 14,400
To know more please follow the following page.
The pricing for the Gemma API can be complex. There is a free tier, which is a great way to start experimenting. For paid tiers, pricing is often based on the number of tokens used (input and output tokens). The specific prices can vary, but generally, for models like Gemma, the cost per million tokens is very low, making it a cost-effective choice for many applications.
Thanks.
i dont want to host the models localy can i just pay per token like i already do with the gemini api with increased limits ? because the rpm is too low for my use case
Yes, you can pay for increased API limits for Gemma by upgrading to a paid tier of the Gemini API. The free tier is designed for initial testing and development for multiple services.
i don’t think that is the case because on Rate limits | Gemini API | Google AI for Developers all tiers have the same limits for gemma models
Yes, If you are particularly looking for only rate limit related service that’s same across all tiers for Gemma models.