Skip to content

74 listings across 43 models

Data policies

from the provider's published policies

Headquarters: USPrivacy policyTerms of serviceStatus page

On training

For clarity, Groq is not permitted to use Inputs or Outputs for training or fine-tuning any AI Model Services or other models, unless explicitly granted permission or instructed by Customer.

Source: console.groq.com/docs/legal/services-agreement

On keeping your data

By default, Groq does not retain customer data for inference requests

Possible sources (2)
More from Groq on keeping your data (2)

All customers may enable Zero Data Retention (ZDR) in Data Controls settings.

Possible sources (2)

These logs are retained for up to 30 days, unless legally required to retain longer.

Possible sources (2)
On where data is processed

Groq is located in the United States, and maintains processing operations in various global jurisdictions.

Source: Groq privacy policy

We take these statements from Groq's own published policies each time we refresh the data, so they can change or disappear when those pages do. Check with the provider before you rely on any of this.

Page 1 of 2. The lists below split this page's listings into those sold by Groq, run by Groq, or sent to Groq by a marketplace.

Sold by Groq

22 listings on this page sold by Groq.

Default order. Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
Qwen 3.8 27B
$0.80$4.00131KNot measured
Qwen3.6 27B
$0.60cache read $0.30$3.00131KNot measured
Kimi K2 Instruct
$1.00cache read $0.50$3.00131KNot measuredNot measured
Kimi K2 Instructmoonshotai/kimi-k2-instruct-0905 · 0905
$1.00$3.00Not published by the providerNot measuredNot measured
GPT OSS 120B
$0.15cache read $0.075$0.60131K
Qwen3 32B
$0.29$0.59Not published by the providerNot measured
$0.75$0.99131KNot measuredNot measured
Llama 4 Maverickmeta-llama/llama-4-maverick-17b-128e-instruct
$0.20$0.60131K
GPT OSS 20b
$0.075cache read $0.0375$0.30131K
Llama 3.1 405Bllama-3.1-405b-reasoning · reasoning
Not enough data yet$0.59$0.79Not published by the providerNot enough data yetNot measured
Llama 4 Scoutmeta-llama/llama-4-scout-17b-16e-instruct
$0.11$0.34Not published by the providerNot measured
Gemma 2 9B IT
$0.20$0.20Not published by the providerNot enough data yetNot measuredNot measured
Gemma 7B IT
$0.07$0.07Not published by the providerNot enough data yetNot measuredNot measured
Llama Prompt Guard 2 22M
Not enough data yetNot enough data yet$0.03$0.03512Not enough data yetNot measuredNot measured
Llama 3.2 1B Preview
Not enough data yetNot enough data yet$0.04$0.04Not published by the providerNot enough data yetNot measuredNot measured
Llama Prompt Guard 2 86m
Not enough data yetNot enough data yet$0.04$0.04512Not enough data yetNot measuredNot measured
Llama 3.1 8B Instant
Not enough data yetNot enough data yet$0.05$0.08131KNot enough data yetNot measuredNot measured
Llama3 8B 8192
Not enough data yetNot enough data yet$0.05$0.08Not published by the providerNot enough data yetNot measuredNot measured
Llama 3.2 3B Preview
Not enough data yetNot enough data yet$0.06$0.06Not published by the providerNot enough data yetNot measuredNot measured
GPT OSS Safeguard 20B
Not enough data yetNot enough data yet$0.075cache read $0.0375$0.30131KNot measuredNot measured
llama-3.2-11b-text-preview
Not enough data yetNot enough data yet$0.18$0.18Not published by the providerNot enough data yetNot measuredNot measured
Not enough data yetNot enough data yet$0.18$0.18Not published by the providerNot enough data yetNot measuredNot measured

Run by Groq

28 listings on this page where Groq runs the model. Another provider sells these listings.

Default order. Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
MiniMax M2.7
 Hugging Face$0.60$1.80197KNot measured
GPT OSS 120B
 Eden AI$0.15cache read $0.075$0.60131K
Kimi K2moonshotai/kimi-k2-instruct
 Llmgateway$1.00cache read $0.50$3.00Not published by the providerNot measured
Qwen3 32B
 Hugging Face$0.29$0.59131KNot measured
GPT OSS 20b
 Eden AI$0.075cache read $0.0375$0.30131K
Llama 3.3 70B Instruct
 Hugging Face$0.59$0.79131KNot measured
Llama 4 Scoutmeta-llama/Llama-4-Scout-17B-16E-Instruct
 Hugging Face$0.11$0.34131KNot measured
Llama 3.1 8B Instruct
 Hugging Face$0.05$0.08131KNot measuredNot measured
Llama Prompt Guard 2 22M
 Eden AINot enough data yetNot enough data yet$0.03cache read $0.015$0.03512Not enough data yetNot measuredNot measured
Llama Prompt Guard 2 86m
 Eden AINot enough data yetNot enough data yet$0.04cache read $0.02$0.04512Not enough data yetNot measuredNot measured
GPT OSS Safeguard 20B
 Eden AINot enough data yetNot enough data yet$0.075cache read $0.0375$0.30131KNot measuredNot measured
Show all 28 listings

Showing 11 of 28: one listing for each model.

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)

How this data works