Groq vs Google Gemini API: free tiers compared

Both columns come from each provider’s own pages, with the day each line was read. Where a page does not state a value, the cell says so.

Groq and Google Gemini API free tiers side by side
Line	GroqFree tier	Google Gemini APIFree tier
Free allowance	30 req/min · 1,000 req/day Chat models; 8K tokens/min and 200K tokens/day. 3 Oct	Not published Limits are per project and shown only inside Google AI Studio. 3 Oct
Requests / min	30 openai/gpt-oss-120b, openai/gpt-oss-20b, qwen/qwen3.8-27b. 3 Oct	Not published Not published per model: “Rate limits … can be viewed in Google AI Studio.” 3 Oct
Requests / day	1,000 Chat models. Prompt-guard models: 14,400/day. 3 Oct	Not published Not published per model: “Rate limits … can be viewed in Google AI Studio.” 3 Oct
Tokens / min	8,000 3 Oct	Not published Not published per model. 3 Oct
Tokens / day	200K 3 Oct	Not published Not published per model. 3 Oct
Card to start	Not published The rate-limits page does not say whether a card is needed to sign up. 3 Oct	Not required The Free tier qualifies with an “Active project or free trial”; no billing tier is involved. 3 Oct
Trains on your data	Not published The data page says “By default, Groq does not retain customer data for inference requests”. It does not address training. 3 Oct	Yes On unpaid services, content is used “to provide, improve, and develop Google products”, and human reviewers may read it. 3 Oct
Commercial use	Not verified	Restricted Only Paid Services may be used for API clients offered to users in the EEA, Switzerland or the UK. 3 Oct
OpenAI SDK	Yes No logprobs, logit_bias or n > 1. 3 Oct	Yes 3 Oct
Image input	Yes qwen/qwen3.8-27b; up to 3 images per request. 3 Oct	Yes Gemini models take images natively. 3 Oct

22 of 24 cells above were read on the provider’s own page in the last 30 days (oldest check: 3 Oct 2026). Each check mark opens that page. The other 2 say “not verified” instead of repeating a number from someone else’s article.

Where they differ

Groq publishes its requests per minute (30); Google Gemini API does not publish that figure.
Groq publishes its requests per day (1,000); Google Gemini API does not publish that figure.
Google Gemini API commercial use: restricted. Only Paid Services may be used for API clients offered to users in the EEA, Switzerland or the UK.

Only differences confirmed on both sides are listed. We do not rank speed or output quality: this site does not measure them.

Same code for both

Both expose an OpenAI-compatible endpoint. Choose either provider below and compare the three values that change.

base_url: https://api.groq.com/openai/v1
model: openai/gpt-oss-120b
key from: $GROQ_API_KEY

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.groq.com/openai/v1",
    api_key=os.environ["GROQ_API_KEY"],
)

reply = client.chat.completions.create(
    model="openai/gpt-oss-120b",
    messages=[{"role": "user", "content": "Reply with the word: ok"}],
)
print(reply.choices[0].message.content)

Endpoint read on 3 Oct 2026· console.groq.com

The key is read from an environment variable on your machine. This site has no field for keys and never asks for one.

Other comparisons

Questions and answers

Which has the higher free limit, Groq or Google Gemini API?

Groq: 30 req/min · 1,000 req/day. Google Gemini API: not published (Limits are per project and shown only inside Google AI Studio). The units differ between providers, so compare them against your own traffic.

Do Groq and Google Gemini API need a credit card?

Groq: not published (the rate-limits page does not say whether a card is needed to sign up). Google Gemini API: not required.

Which one trains on my prompts?

Groq: not published (the data page says “by default, groq does not retain customer data for inference requests”. it does not address training). Google Gemini API: yes.

Can I use both with the same code?

Yes. Both expose an OpenAI-compatible endpoint: https://api.groq.com/openai/v1 and https://generativelanguage.googleapis.com/v1beta/openai/. Only the base URL, the model name and the key change.