Model inference platform with per-token Model APIs and dedicated GPU deployments billed per minute. Basic plan is pay as you go at $0 per month.
Teams deploying open-weight or custom models to production who want dedicated GPU deployments billed per minute of active compute, plus per-token APIs for hosted models.
Teams that need a published price for Pro or Enterprise: neither has a list price and both require a quote. Also not for buyers who need to know the free credit amount up front, since the page says new accounts get credits without stating how much.
Pros
Cons
Dedicated deployments are billed per minute of active compute. Listed rates include $0.10833 per minute for an H100 80 GiB, $0.16633 for a B200 180 GiB, $0.06667 for an A100 80 GiB and $0.02012 for an A10G. The page shows $ without a currency code.
No. The Basic plan is $0 per month pay as you go, but Pro and Enterprise show no list price. The page offers volume discounts and tells you to get a quote, so you must contact Baseten for those figures.
Model APIs are priced per 1M tokens. Examples: Kimi K3 is $3.00 input, $0.30 cached input and $15.00 output; GPT OSS 120B is $0.10 input and $0.50 output. The page does not name the currency beyond the $ sign.
The pricing page says idle time is not billed. You pay for active compute, including deploying, scaling and predictions. Check how your deployment's scale-to-zero settings behave before assuming a zero bill, since the page does not detail cold start costs.
The page says new accounts receive credits but does not state the amount. Basic is $0 per month with pay as you go, so ask the vendor or check the signup flow for the current credit figure.
Facts sourced from the vendor’s own site, last verified 2026-10-30.
The Weekly Source
One email a week. New listings, repriced tools, and the ones we would skip - including this one if its pricing changes.
