Provider-neutral comparison

LLM inference pricing grid

Compare direct, OpenRouter, OrcaRouter, and Vercel AI Gateway offers while keeping model ownership and the actual inference provider clear.

Current public offers

Prices are USD per million tokens. Models and their default offers start with the highest cost performance; open the bar below each model to compare every matching route.

Loading offers…

Columns

Visible columns

Monthly cost TPS override

Applies one throughput value to every monthly cost.

Scroll horizontally to see every price column.

LLM inference offers by model and creator, route and provider, model size and quantization, throughput, token price, cost performance, intelligence, and estimated monthly output cost.
Loading…
Loading verified offers…

Monthly assumes the listed TPS can be sustained at maximum output for 2,592,000 seconds (a 30-day month). It excludes input-token cost and is an estimate, not a capacity quote.

Throughput is an observed provider benchmark where available. Workload, region, load, and routing overhead can materially change real-world speed; hover or focus a value for its methodology note.