Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Cost levers - verified

How to cut your Groq API bill

Prompt caching: automatic. Cache lifetime: 2 hours without use. Batch: 50% off, 24h to 7d. Batch and caching stack: no. Service tiers: Flex (same price, 10x rate limits).

Read from console.groq.com. Compare every provider's cost levers

Caching
Automatic
Cache hit saves
See docs
Batch
50% off
Service tiers
1

Prompt caching

Mode

Automatic: repeated prefixes are cached with no code change

Automatic prefix caching, cannot be disabled; currently only GPT-OSS 20B, GPT-OSS 120B and GPT-OSS-Safeguard 20B.

Cache lifetime

Expires after 2 hours without use

Minimum prompt

128 to 1024 tokens depending on the model

Cache write cost

No additional fees

Cache read cost

50% discount on cached input tokens

Storage cost
Not stated
How to enable

Automatic; put static content first.

From the docs
“Prompt caching works automatically on all your API requests with no code changes required and no additional fees.”
console.groq.com
“Automatic Expiration: All cached data automatically expires after 2 hours without use.”
console.groq.com
“Minimum Prompt Length: The minimum cacheable prompt length varies by model, ranging from 128 to 1024 tokens depending on the specific model used”
console.groq.com
“Prompt caching is provided at no additional cost. There is a 50% discount for cached input tokens.”
console.groq.com
“Prompt caching is currently only supported for the following models:”
console.groq.com
“Cached tokens do not count toward your rate limits.”
console.groq.com

Batch API

Discount

50% off

50% off all tokens; does not stack with the caching discount

Window

Processing window settable from 24h to 7d; incomplete jobs expire (charged only for completed requests)

Limits

JSONL up to 50,000 lines and 200MB per file; unlimited number of batch jobs; separate rate limits; files kept up to 30 days

Endpoint

https://api.groq.com/openai/v1/batches (lines target /v1/chat/completions, /v1/audio/transcriptions, /v1/audio/translations)

Stacks with caching

No: you get one discount or the other, not both

“Note: The batch discount does not stack with prompt caching discounts. All batch tokens are billed at the 50% batch rate regardless of cache status.”
console.groq.com
From the docs
“Batch processing lets you run thousands of API requests at scale by submitting your workload as an asynchronous batch of requests to Groq with 50% lower cost, no impact to your standard rate limits, and 24-hour to 7 day processing window.”
console.groq.com
“Note: The batch discount does not stack with prompt caching discounts. All batch tokens are billed at the 50% batch rate regardless of cache status.”
console.groq.com
“The completion window for batch jobs can be set from to 24 hours (`24h`) to 7 days (`7d`).”
console.groq.com
“The Files API supports JSONL files up to 50,000 lines and 200MB in size.”
console.groq.com
“You are only charged for successfully completed requests”
console.groq.com

Service tiers

Flex trades speed or availability for a lower price; Priority and Fast cost more for lower latency or reserved capacity.

Flex

Flex Processing

Set with service_tier: flex

Price effect

Same price as on-demand

Trade-off

10x higher rate limits than on-demand (paid customers); fails fast with status 498 capacity_exceeded when flex capacity is unavailable

From the docs
“This tier offers significantly higher rate limits while maintaining the same pricing as on-demand processing.”
console.groq.com
“Flex processing is available for all models to paid customers only with 10x higher rate limits compared to on-demand processing. Pricing matches the on-demand tier.”
console.groq.com

Other discounts and surcharges

Other discounts on the pages checked: Not stated

Groq per-model prices

This site does not track per-model Groq prices. Use the sources below for the current rate card.

Research notes

Flex is a throughput lever, not a price discount.

FAQ

Does Groq cache prompts automatically?

Automatic: repeated prefixes are cached with no code change. Automatic prefix caching, cannot be disabled; currently only GPT-OSS 20B, GPT-OSS 120B and GPT-OSS-Safeguard 20B. How to enable: Automatic; put static content first.

How much does a cache hit cost on Groq?

50% discount on cached input tokens. This site has no per-model Groq prices to compute it from.

Does Groq have a batch API discount?

50% off all tokens; does not stack with the caching discount. Window: Processing window settable from 24h to 7d; incomplete jobs expire (charged only for completed requests). Stacks with caching: no: you get one discount or the other, not both.

What service tiers does Groq offer?

Flex: Same price as on-demand.

Sources

Last verified from Groq's own pages below. Quotes are verbatim; each was matched against the live page.

Newsletter

Know when a provider changes its prices

One email per week. New model releases, pricing changes and what they mean for what you build.