OpenModelsHQ

Groq

Groq serves 5 open models from 3 labs on OpenRouter, in 5 offers. Of the 4 models it shares with other providers, it has the lowest blended price on 0. It is the only provider for 1 model. Its median blended price is $0.2625 (₹25) per million tokens, blended 3:1 input to output. It does not list a precision for its offers.

Models served
5
3 labs · 5 offers
Median blended / 1M
$0.2625
₹25 · 3:1 input to output
Lowest price, shared models
0%
0 of 4 · only provider for 1
Max context
196,608
Not listed 5

Company and policies

Headquarters listed by OpenRouter
United States
Datacenter countries listed by OpenRouter
Not listed

Provider-level locations do not guarantee where a particular request is processed. Check the provider's terms for data handling. Directory source, checked 14 Nov 2026.

Precision and price rank

Precision mix
Precision as the provider lists it on OpenRouter. 4-bit covers FP4, NVFP4, MXFP4 and INT4; 16-bit covers BF16 and FP16.
Precision mix: data
OffersNot listedTotal
Offers55
Price rank on shared models
One entry per model that two or more providers serve, using Groq's cheapest offer. A tie for the lowest price counts as lowest. Not shown: 1 model with no other provider.
Price rank on shared models: data
RankModels
Lowest or tied0
2nd1
3rd0
4th or lower3

Every model Groq serves 5

USD and INR per million tokens
PrecisionLabPricing
$0.2625₹25$0.15₹14$0.60₹5715 of 22131,072Not listedOpenAIAll providers
$0.1312₹12$0.075₹7.07$0.30₹2810 of 10131,072Not listedOpenAIAll providers
$0.1312₹12$0.075₹7.07$0.30₹281 of 1Only provider131,072Not listedOpenAIAll providers
$0.64₹60$0.59₹56$0.79₹744 of 8131,072Not listedMetaAll providers
$0.90₹85$0.60₹57$1.80₹1702 of 2196,608Not listedMiniMaxAll providers
Rank is this offer's position among the model's providers by blended price (3 parts input to 1 part output). 1 is the lowest. Equal prices are ordered by provider name and marked "Ties for lowest".

Where Groq is cheapest

0 of 4 models shared with other providers

Groq does not have the lowest blended price on any model that other providers also serve.

Only provider for 1 model: gpt-oss-safeguard-20b. With no other provider to compare, these are not counted as cheapest.

Groq questions

Which open models does Groq serve?

Groq serves 5 open models from 3 labs on OpenRouter. The most downloaded on Hugging Face are gpt-oss-20b, gpt-oss-120b and Llama-3.3-70B-Instruct. Prices were fetched 14 Nov 2026.

Is Groq the cheapest provider for any open model?

No. Its best position is 2nd of 2 providers, for MiniMax-M2.7 at $0.90 (₹85) per million tokens, blended 3:1 input to output. It is also the only provider for 1 model, which we do not count as cheapest.

What precision does Groq serve models at?

Groq does not list a precision for any of its 5 offers on OpenRouter.

About this data

Every number on this page comes from the endpoints OpenRouter lists for each model, fetched 14 Nov 2026. INR is shown beside USD at ₹94.29 per $1. Prices change, so check the provider before you build on a number.