Sentry migrates internal AI usage to OpenRouter
Understand this story
In July 2026, Sentry migrated its internal AI usage to OpenRouter, cleaned up about 500 Anthropic and OpenAI keys within three weeks, shut down direct access to both companies in August, and reduced the runaway bill cap by about 93%. After the migration, more than 45 applications call models through a single account, and spending can be traced to specific people or workspaces; engineers can try new models without vendor approval, the account used 147 models in total in the first three months, and switching models only requires changing configuration. Sentry requires zero data retention, so this setting is enabled by default; otherwise routing is restricted to its own vendor keys only through bring-your-own-key (BYOK). Regarding permissions, everyone including teams outside engineering gained access on day one and can create any number of keys themselves. Each workspace has an overall budget cap approved by Sentry's CFO, and no workspace has currently hit the cap; the default workspace also sets a budget of $1,500 per person per week.
AI overview based on source reports · Updated 1 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- OpenRouter AnnouncementsHow Sentry Capped Runaway AI Bills Without Restricting Access
Sentry moved its internal AI usage onto OpenRouter in July 2026, retiring roughly 500 Anthropic and OpenAI keys over about three weeks and switching off direct access to both providers in August. Keys are bound to users or workspaces so spend can be traced to an owner and model, and per-person and per-workspace budgets cap runaway bills. Sentry says the change gave it tighter control and wider model access at the same time, with 45+ apps and 147 models used in the first three months.
Activity for this story
Not enough continuous observations to show a trend yet.