AI models

Groq vs OpenRouter vs Google Gemini

Groq, OpenRouter, Google Gemini side by side: price, free tier, who each one fits and when to skip it.

Item	Groq	OpenRouter	Google Gemini
What it does	Very fast inference API for open models such as gpt-oss, running on Groq's own chips.	One OpenAI-compatible API and one bill for 500+ models from 80+ providers.	API for Google's Gemini models, which read text, images, audio and video, with a free tier.
Best for	A WhatsApp bot or voice assistant where every second of delay loses the user and an open model is good enough.	A SaaS that wants to switch between Claude, GPT and Gemini by changing one string, with a single credit balance.	A SaaS that processes lots of PDFs, audio or video and wants to prototype on the free tier before paying.
Skip it if	You need the top-tier reasoning of Claude, GPT or Gemini, since Groq only serves open models.	You run high volume on a single model, where the 5.5% fee on top of the provider price adds up.	You need prices locked in for next year: Gemini 4 Argon doubles on Dec 1, 2026 and the 3.x Flash models double on Jan 1, 2027.
Setup	Needs a developer. Minutes for a free key; an afternoon for the first version in your app.	Needs a developer. About an hour if your app already calls OpenAI; an afternoon from scratch.	Needs a developer. Minutes to get a free key; an afternoon for the first version in your app.
Price	gpt-oss-120b: $0.15 input / $0.60 output per million tokens	Provider price plus a 5.5% platform fee	Gemini 4 Argon: $2 input / $10 output per million tokens (rising Dec 1, 2026)
What it costs in practice	Estimate: 1,000 users × 30 chats a month × about 2,000 tokens per chat (1,500 in, 500 out) = 45M input and 15M output tokens. On gpt-oss-120b ($0.15 in / $0.60 out per million): 45 × $0.15 + 15 × $0.60 ≈ $16/month. The bill is small, but the base limit of 200,000 tokens a day (about 6M a month) covers a tenth of this volume, so check your real limits first.	Estimate: 1,000 users × 30 chats a month × about 2,000 tokens per chat (1,500 in, 500 out) = 45M input and 15M output tokens. On Claude Sonnet 5.5 at Anthropic's list price ($2 / $10 per million) that's about $240 in usage, plus 5.5% on the credits you buy: about $253/month. On a cheap model like gpt-6-luna it's about $12 + fee ≈ $13.	Estimate: 1,000 users × 30 chats a month × about 2,000 tokens per chat (1,500 in, 500 out) = 45M input and 15M output tokens. On Gemini 3.8 Flash ($0.75 / $3.75 per million): about $90/month today and $180 from January 2027. On Gemini 4 Argon: about $240 today and $480 from December 1, 2026, which is above Tier 1's $250 cap.
Free plan	Free plan with per-model rate limits per minute and per day.	25+ free models, 50 requests a day (1,000 after buying $10 in credits).	Free tier with rate limits on the Flash and Flash-Lite models; Gemini 4 Argon is paid only.
Price per million tokens (main model)	gpt-oss-120b: $0.15 in / $0.60 out	Same as the provider + 5.5% fee	Gemini 4 Argon: $2 in / $10 out
Models	Open models: Llama, gpt-oss, Whisper	500+ models, 80+ providers	Gemini 4, Gemini 3.x Flash, Nano Banana
Free tier	Rate-limited free plan	50 requests/day on free models	Rate-limited free tier on Flash models
Open source	No: no	No: no	No: no
Before you build	The catalog is short. Production LLMs are Llama 3.1 8B, Llama 3.3 70B and gpt-oss 120B/20B, and the Llama ones say "contact sales" for price. Models like Qwen are preview only. Preview models should not be used in production, because they may be discontinued at short notice. Base limits are tight: the docs list gpt-oss-120b at 30 requests and 8,000 tokens per minute, and 200,000 tokens per day, per organization. A few simultaneous users can hit that; higher limits are for select workloads. Payment is automatic from a credit card, US bank account or SEPA debit. If a payment fails and isn't fixed, the service may be suspended.	The 5.5% fee is charged when you buy credits and is non-refundable. You can only refund unused credits within 24 hours of the purchase, and crypto payments never. OpenRouter reserves the right to expire unused credits one year after purchase. By default, the same open model can be served by different providers, picked mostly by price, and providers may run it at different precision (quantization). Answers can vary between calls unless you pin providers. Routing allows providers that collect data by default. If you handle customer data, set data_collection to "deny" so only providers that don't keep it are used. Free models are capped at 20 requests a minute and 50 a day (1,000 after $10 in credits). Failed attempts count, and a negative balance blocks them.	On the free tier, Google uses what you send to improve its products and human reviewers may read it. The terms say not to send sensitive, confidential or personal information. Paid usage isn't used this way. Prices are going up: Gemini 4 Argon goes from $2/$10 to $4/$20 per million tokens on December 1, 2026, and Gemini 3.8 and 3.6 Flash double on January 1, 2027. Tier 1 has a $250 monthly spend cap. Reaching Tier 2 ($2,000 cap) takes $100 in paid spend and 3 days after the first payment; at the cap, the API pauses until the 1st of next month. Prepaid credits expire 12 months after purchase and are generally non-refundable. When the balance hits $0, every linked key stops working until you add credits. Google Cloud welcome credits generally can't be used for the Gemini API, especially on accounts opened after March 2, 2026.
Price checked	Checked on Oct 28, 2026 · official pricing	Checked on Oct 28, 2026 · official pricing	Checked on Oct 28, 2026 · official pricing

Prices are read from each official pricing page on the date shown and can change. Always confirm before signing up.

Popular comparisons

Anthropic vs OpenAI vs Google Gemini

Faundr

Picked your tools? Build it with AI, safely.

Faundr keeps your project’s memory, blocks what can’t be undone and tells you if it’s safe to launch, inside Claude Code and Codex.

Get started free