Coding & apps
GroqCloud
GroqCloud is Groq's hosted inference platform with an OpenAI-compatible API. The Free plan needs no card and lets you call 10 models, including GPT-OSS 120B, Qwen3.8-27B and Whisper, each with its own free rate limits.
Free plan
No card required. Limits apply per organization, and cached tokens do not count toward them.
Developers who want to try open-weight LLMs, speech-to-text and text-to-speech APIs for free before paying.
Plans & allowances
Official plan pricingFree plan
View planAllowance
Per-model rate limits, counted per organization
Features
- Free plan
- Allowance reset
Eligibility & billing details
Availability
Regional coverage is not specified.
Plan details
No card required. Limits apply per organization, and cached tokens do not count toward them.
Developer plan
View planAllowance
Higher rate limits, plus Batch and Flex processing
Features
- Paid plan
- Allowance reset
Eligibility & billing details
Availability
Regional coverage is not specified.
Plan details
Requires a credit card, US bank account or SEPA debit account and is billed by usage. New accounts are invoiced early as cumulative usage reaches $1, $10, $100, $500 and $1,000, then monthly; thresholds differ in India.
Official sources
Checked 2026-10-31