Groq

Ultra-low-latency inference (LPU).

Update, Nov 12, 2026:Free and Developer plans, billed per token (for example GPT OSS 120B at $0.15 input and $0.60 output per 1M tokens).

Rank
#19 of 66#8 of 17 in Infrastructure
Adoption score
82 / 1001.5 points above the category average
Entry price
UsageConfirmed Nov 12, 2026
Open source
NoHosted by the vendor

Our assessment

Groq ranks #8 of 17 in Infrastructure and #19 of 66 overall, with an adoption score of 82, 1.5 points above its category average. Built by Groq, it delivers ultra-low-latency inference (LPU).

It is proprietary, with entry pricing at Usage. We confirmed that price on the vendor's own pricing page.

Where it fits: Blazing fast LPU inference for open-weights models like Llama 3.

Prices and facts come from Groq's own pages; the full list is on the sources page. Suggest an edit.

Best for
Blazing fast LPU inference for open-weights models like Llama 3
Watch out for
Proprietary and cloud-managed; entry price Usage.
Consider instead
Ollama (Free) · vLLM (Free)

How it compares in Infrastructure

History

Tracked since Oct 13, 2026.

SnapshotRankAdoptionEntry pricePrice check
Nov 20, 2026#19 same82UsageConfirmed
Nov 12, 2026#19 same82UsageConfirmed
Oct 29, 2026#19 same82UsageNot yet
Oct 15, 2026#19 same82UsageNot yet
Oct 13, 2026#19 82UsageNot yet

Key facts

CategoryInfrastructure
MakerGroq
LicenseProprietary
Open sourceNo
Self-hostNo
Entry priceUsage
Category rank#8 of 17
Price checkConfirmed on the vendor's own page

Compare Groq

Get an email when the index changes

Price changes, new tools, acquisitions and shutdowns, sent when the data is updated.