Tidal Telemetry
Everything else on this site is people doing work on your bill. Tidal Telemetry is software you sign up for and run yourself. It is free during beta, it requires no engagement with us, and this page exists so the difference is not confusing.
What it is
Hosted LLM observability. Install the SDK, call the init function, and every model call your application makes starts arriving as a trace: model, tokens in and out, an estimated cost per call, latency, errors.
- One line, roughly 50 libraries. Auto-instrumentation covers OpenAI, Anthropic, Mistral, Bedrock, LangChain, LlamaIndex, vector databases and the common agent frameworks. Python and Node get the one-line install.
- Plain OpenTelemetry, so the SDK is optional. Ingest is OTLP over HTTPS with a bearer token. Point any OpenTelemetry exporter at your project endpoint and it behaves identically. The SDK is a thin wrapper over the Apache-licensed OpenLIT SDK.
- Isolated per-project storage. Each project gets its own database and its own ingest key, enforced at the storage layer rather than by a filter in a query.
- A built-in MCP server. Agents can verify the setup, send a test trace and pull usage reports directly, which is a faster loop than reading a dashboard.
What it costs
Nothing, currently. It is in beta and free with no credit card, paid tiers are planned, and beta users keep a free plan when those arrive. That is the actual state of it rather than a launch offer with a countdown attached.
How it relates to the engagement
Telemetry shows you what your LLMs spend. The FinOps LLM engagement is the work that reduces it, priced on verified savings. The two are cross-promoted and deliberately not bundled.
- Use Telemetry and never talk to us. There is no sales gate on it and no engagement prerequisite.
- Hire us and never use Telemetry. Audits run against provider billing data - the OpenAI Costs endpoint, the Bedrock CUR, the Vertex BigQuery export - not against our own product.
- Below the engagement minimum, this is the better answer. Instrument it yourself, read your own numbers, and use the free templates to allocate them.
Keeping the two apart is a constraint we put on ourselves for a reason. A cost consultant who also owns the meter has an obvious problem, so every baseline behind every fee comes from raw provider invoices.
Signup is at telemetry.llmcfo.com, and the setup docs cover Python, Node, plain OpenTelemetry, Docker and Kubernetes. If what you want instead is someone to read your invoice and tell you where the money went, that is the free audit.
FAQ
What is Tidal Telemetry?
A hosted LLM observability product from the same team as FinOps LLM. You install an SDK, call one init function, and every model call arrives as a trace with model, token counts, estimated cost, latency and errors attached.
Do I have to use Tidal Telemetry to work with FinOps LLM?
No, and the reverse is also true. Engagements are measured against your raw provider invoices and whatever tooling you already run, and nothing in our pricing licenses a dashboard to you. You can sign up for Telemetry without ever speaking to us.
What does Tidal Telemetry cost?
It is free during the beta with no credit card. Paid tiers are planned, and beta users keep a free plan when they arrive.
Do I have to install your SDK to send data?
No. Ingest is plain OTLP over HTTPS with a bearer token, so any OpenTelemetry exporter can point at your project endpoint directly. The SDK is a convenience wrapper over the Apache-licensed OpenLIT SDK, not a requirement.