Modal restricts Shared API token-based pricing to Team and Enterprise plans
Modal now limits its token-billed Shared API (launched with Kimi K3) to Team and Enterprise plans; Starter keeps the same models via a per-second Auto Endpoint instead.
Modal's Shared API launch post said the OpenAI-compatible, token-billed endpoint was covered by Starter's standard $30/month free-compute offer, implying access on any plan.
The same post now says the Shared API's token-based pricing is available to Team and Enterprise customers only; Starter and other plans retain access to the same hosted models (e.g. Kimi K3) through a dedicated per-second Auto Endpoint instead.
Modal’s July 29 announcement of an OpenAI-compatible Shared API for Kimi K3 originally framed it as available on any plan, with Starter’s $30/month free-compute credit covering ongoing usage. As of this week, the same blog post has been edited to gate Shared API token-based pricing to Team ($250/mo + compute) and Enterprise customers only. Starter and other lower tiers are not locked out of the underlying model — they can still reach Kimi K3 through Modal’s per-second Auto Endpoint — but lose access to the token-metered pricing surface itself. Modal has not published per-token input/output rates for the Shared API on its pricing page or billing docs.
Modal edited its Kimi K3 / Shared API announcement post to restrict the token-based Shared API to Team and Enterprise customers only, removing earlier language that had implied Starter's $30/month credit covered ongoing Shared API usage on any plan. Starter and other plans keep access to the same hosted models via a dedicated per-second Auto Endpoint instead; no per-second rates or plan fees changed.