Listed in the following categories:
Generative AI Infrastructure
Logo

Groq Costs

(0 reviews)
Cloud platform for fast inference of open language models through a programming interface.

Groq Pricing

Last edited: September 2026
  • Developer
    Pay-Per-Use
    Billing by usage, priced per model.
    GPT-OSS 20B: USD 0.075 per 1M input tokens, USD 0.30 per 1M output tokens
    GPT-OSS 120B: USD 0.15 input, USD 0.60 output
    Qwen3.8 27B: USD 0.80 input, USD 4.00 output
    Whisper Large v3 Turbo: USD 0.04 per hour of audio
    Speech model Orpheus V1 English: USD 22.00 per 1M characters
    Rate limits per model, around 250,000 tokens and 1,000 requests per minute
  • Enterprise
    Upon request
    For models and capacity the provider does not price publicly.
    Everything in Developer, plus:
    Access to models sold through sales only, such as Llama 3.1 8B, Llama 3.3 70B and MiniMax M2.7
    Dedicated capacity on reserved hardware
    Higher rate limits without a published tier
    Enterprise governance with auditability and controls
Item 1 of 2
All prices in USD and billed by usage. The provider publishes no fixed-price tiers, only a price per model. The provider’s model overview prevails.

Reviews for Groq with pricing information (0)

There are no reviews with pricing information yet.

Alternatives to Groq

Item 1 of 8
See all alternatives to Groq