Keep 90% of every dollar you save

Savings are priced against live token prices per model. Every month we measure what you actually spent versus what you would have spent without Tokeven's coaching, using live provider pricing. You keep 90% of every dollar saved. If we save you nothing, you pay nothing.

Every API call that passes through the Tokeven wrapper counts as one tracked call. Batch calls count as one event per request.

Yes. Start on the free plan, install the wrapper, and see your data. When you want savings proven per person and per team, upgrade. There's no contract and you can leave anytime.

Every savings claim links back to the specific calls, models, and recommendations that produced it. You can audit the math in the dashboard or export the raw data via API.

No. Your Anthropic, OpenAI, Google, and xAI API keys never reach Tokeven - the wrapper runs locally using your existing client, and the only credential we hold is a write-only ingest token you can revoke. Prompt content never reaches Tokeven through the ingest pipeline: we receive usage metadata - token counts, model and provider IDs, latency, and labels you choose, such as a project or an optional name for who ran the call - and nothing read out of your prompts. Two opt-in tools are the disclosed exceptions, and neither touches your instrumented traffic - Prompt Revision sends text you paste to Anthropic to rewrite it, and advisor.improve() sends the prompt you are improving to Anthropic using your own key, bypassing our servers entirely. See /privacy §2.4.

The free tier helps any individual developer, regardless of spend. Team pricing kicks in only when we prove savings, so there's no minimum to start.