Book a demo with our team

What the demo covers

We show you telemetry.dev with the frameworks and model providers that you use. Name them in the message field, and the walkthrough uses your stack.

We start with tracing. A trace groups the instrumented model calls, tool steps, and retrievals in one request or agent run. We open a trace and look at its token usage, cost, latency, and errors.

Then we look at cost. telemetry.dev calculates cost at ingest from reported token usage and its model-pricing catalog. Cache-read and cache-write tokens use separate rates when the catalog has them. A cost that your span sets replaces the calculated cost.

Last, we look at privacy controls. Each project has its own settings for prompt capture, redaction, and retention. You can turn off storage of recognized prompt and completion fields for each environment. This does not stop transmission or remove other content from logs, errors, or custom attributes. You can add your own redaction patterns. Retention is 7, 14, 30, or 90 days, up to the limit of your plan.

After you send the form

The form sends your name, email, company, and message to our team. We reply to the email address that you enter. The message is optional and can have up to 1,000 characters.

If you prefer to write first, email support@telemetry.dev.

Start without a call

You do not need a demo to use telemetry.dev. The Free plan costs $0 and needs no credit card. It includes 30,000 ingestion units per month, 14-day retention, one project, and two seats. Each span, log record, or metric data point uses one unit. Create a free account or compare the paid plans on the pricing page.

The quickstart shows how to send your first trace with TypeScript or Python. The integrations page has a setup guide for each provider, framework, coding agent, and SDK. Other languages can send standard OTLP over HTTP.