Modal added a new Inkling model to its shared endpoints

Modal added a new model, Inkling 2 Small, to its shared endpoints rate card. Per Modal's pricing page, saved Nov 13, 2026, it is priced at $0.30 per 1M prompt tokens, $1.00 per 1M completion tokens, and $0.03 per 1M cached tokens.

The addition is Modal's first pricing-page change since it dropped Notebooks from its Sandbox pricing section on Oct 26, 2026, according to Pulse's record — 18 days earlier. Modal does not state a reason for adding this particular model.

For product and platform teams buying inference by the token, Modal's shared endpoints table is a quick way to compare cost per model, and one more entry widens that shelf without touching rates on anything you're already running. If you're pricing your own AI features by usage, the pattern here is familiar: capacity grows by adding metered options to an existing table rather than renegotiating the table itself.

This is also the kind of move Modal makes more than any other: a catalog change to its usage-metered rate card rather than a change to its own prices. For a team evaluating Modal against other inference providers, that steady drizzle of added models matters more for total cost than any single line in the table — it signals Modal is trying to be a shared endpoint for third-party models as much as a compute host. Watch whether Modal starts bundling discounted rates across its own and third-party models, which would be the next step beyond simply listing more options side by side.