Groq vs OpenRouter: one fast inference provider or a router?
Updated 2026-11-01 · listed A to Z, not ranked
Groq runs models on its own inference hardware (GroqCloud) and sells access per token. OpenRouter is a router: one API in front of many providers. They are not exclusive; a router can send requests to a fast provider.
Listed A to Z, facts from console.groq.com/docs/models and openrouter.ai/pricing on Nov 1, 2026.
- What it is
- Inference provider running models on its own LPU hardware
- Example prices
- GPT OSS 120B $0.15 in / $0.60 out per 1M tokens; GPT OSS 20B $0.075 / $0.30; Whisper Large V3 Turbo $0.04 per hour of audio; some models “contact sales”
- Rate limits
- Per model on the Developer plan, e.g. 250K tokens/min and 1K requests/min for GPT OSS 120B
- What it is
- Router API for 600+ models from 80+ providers
- Example prices
- Each model at its provider’s list price, plus 5.5% (Standard) or 8% (Business) on credit purchases
- Rate limits
- No plan-based limit on paid models; provider limits pass through
How we checked: Per-token prices and limits are quoted from GroqCloud’s models page (Developer plan) and OpenRouter’s pricing page on Nov 1, 2026. Model catalogs change often: confirm with each provider.
How they differ
- Catalog. Groq serves the models it hosts (production and preview lists on its docs). OpenRouter reaches many providers, including closed models.
- Speed. Groq publishes token speeds per model (for example about 500 tokens/second for GPT OSS 120B). On OpenRouter, latency depends on which provider serves the request; pin a provider for consistency.
- Billing. Groq bills per token for its models. OpenRouter adds its fee on credit purchases on top of list prices.
FAQ
How much does Groq cost?
GroqCloud lists per-token prices per model, for example GPT OSS 120B at $0.15 per 1M input tokens and $0.60 per 1M output tokens, and Whisper Large V3 Turbo at $0.04 per hour of audio (checked Nov 1, 2026).
Can I reach Groq-hosted models through OpenRouter?
OpenRouter routes to many providers; check the model’s page on openrouter.ai to see which providers serve it and at what price.
Sources (official pages, read 2026-11-01)
Spot something out of date? Email hello@youraitools.co and we’ll fix it.