Groq Pricing: Plans & API Token Cost Calculator 2026
Free tier plus usage-based per-token rates among the lowest for hosted inference. You trade a narrower model catalog for the speed.
Groq plans and pricing
Free
A rate-limited free tier with access to the full open-weight catalog, enough for development and low-traffic apps
Developer
Enterprise
Groq pricing: the quick answer
Groq bills pure pay-as-you-go token rates as of July 8, 2026, with no subscription tiers and a free plan for testing. Free gets limited usage, Developer is usage-based at 10x higher rate limits, Enterprise is custom with dedicated capacity. From the catalog, Llama 4 Scout runs a fraction of a cent per 1M tokens, GPT-OSS 120B runs $0.15 in and $0.60 out per 1M, and batch jobs take 50% off. No flat fee to plan around, so your bill is whatever the rate card and your volume add up to.
- FreeFree
- DeveloperCustom
- EnterpriseCustom
Groq is free to start, against a $8.50/mo median across 10 large language models tools we track.
API Token Pricing
GroqCloud on-demand rates, groq.com/pricing, read 2026-07-22
Groq cost calculator
Groq Hidden Costs & Pitfalls
There is no flat subscription to budget against here. The cost is the per-model, per-tool rate card, and it runs wider than most buyers expect.
Groq pricing, read against its live plans and category
Positioning
Groq skips the subscription entirely: a free tier plus pay-as-you-go token pricing. Where the category median sits at $8.4/mo, Groq's Developer and Enterprise tiers bill by usage rather than a flat monthly fee. For agentic workloads that make a lot of fast calls, per-token pricing on cheap open models is efficient, and the platform is worth it precisely when your priority is raw inference speed and low token cost rather than model breadth.
Cost drivers
- 1Pricing is straightforward; no documented hidden fees or overage traps found.
Watch-outs
Groq's inference is cheap and fast, but leaning on the free tier for production is risky: strict rate limits and latency that spikes under load. Users report speeds that are generally unmatched alongside consistency that can swing hard when traffic is heavy.
Strengths
The free tier already delivers the headline capability: the fastest inference speeds on the market, at no cost.
- World's fastest inference speed (500+ tokens/sec)
- Custom LPU hardware eliminates sequential processing bottlenecks
- OpenAI-compatible API for near drop-in integration
What users say
“I built a Study OS with Llama 4 + Groq because Otter was too expensive.”
“I tested the playground inference on their website. Insane speeds.”
Editor’s take
Individual developers and startups should start on the Free tier to test the API, then move to the Developer tier for higher token limits and the Flex service tier as production scales. For dedicated capacity and support, the custom Enterprise tier is the path. If you want a predictable flat-rate plan with built-in frontier models instead, Google Gemini at $20/mo covers it.
Oleh KemFounder & Lead AnalystGroq price history
Cheaper Large Language Models tools
Frequently asked questions
Sources & verification
| Source | What was checked | Last checked |
|---|---|---|
| Official Pricing Page | Source of verified tiers | July 8, 2026 |
| Official Website | Official vendor website | — |
| PeerSpot | PeerSpot enterprise peer reviews | — |
Every fact on this Groq pricing page is tied to a named source and a verification date. Freshness-sensitive figures trace to the sources above; verify against the vendor before relying on them.
Explore Groq
Every page on Groq in one place, you are on pricing.
Snapshot, score and verdict
How to get API access, limits, SDKs and what it costs
Latency, throughput, uptime and behaviour under scale
You are here
Hidden fees, discounts and how to negotiate
Compared and ranked vs peers
Price and feature change history
Browse the full Large Language Models category