Skip to main content
This page lists every Tuner feature, what it does and when it’s useful, with a link to the page that covers it. You connect an agent once and define what a good call looks like. That definition scores your production calls and the test calls Tuner places against your agent, so testing needs no second configuration.

New here?

The Quickstart connects an agent and gets your first call evaluated.

Want the vocabulary first?

The glossary defines every term used across the product.

Connecting your agent

The only job of an integration is to get the finished call to Tuner. Once Tuner has it, everything on this page happens on Tuner’s side: transcript, per-turn timing, classification, evals, red flags, and alerts. Which route you take depends only on what you build on, and every route produces the same result once the call lands. Worth knowing: for monitoring, traffic goes one way, from your stack to Tuner over HTTPS. Tuner needs no route into your network, so a fully self-hosted agent connects the same way a managed one does.

Seeing what happened on a call

Worth knowing: the depth of the latency breakdown depends on how you connected. Platforms that hand over a finished transcript can only tell you a call was slow. The SDKs record each stage as it happens, which is what turns “slow” into “the LLM stage added 900ms on turn four”.

Judging every call

This is where you tell Tuner what a good call looks like. Everything defined here runs on production calls and simulated calls alike. Worth knowing: an eval judge can read the tool calls and results and the agent’s own instructions, as well as the transcript. So an eval can check facts, for example “The agent quoted a delivery time the tool never returned”. See What an eval judges. Evals also run in whatever language the call happened in, so a multilingual agent needs one set of evals.

Testing before your customers do

Call Simulation places real voice calls against your agent and scores them with the evals you already configured. It is available on every plan, starting with Pay as You Go. Simulated calls use credits.

Configuring, diagnosing, and automating


Data and compliance controls

Where your agent runs and where Tuner runs are separate questions. Your agent can be entirely self-hosted on any plan. Tuner itself runs as a hosted cloud service, with on-premise deployment available on Enterprise, alongside SOC 2 and HIPAA compliance, custom integrations, and a DPA on request. See the FAQ for both.

How Tuner compares

Useful when you are comparing tools.
In Tuner the evals, outcomes, and intents you define score both simulated and production calls, so you write your criteria once. A behavior you catch in simulation is scored the same way in production, and evals you add after a production failure also generate Pressure Tests on the next run.
Tuner’s SDK integrations record end of utterance, STT, LLM, TTS, and time to first byte for each turn, so you can see which stage was slow. A platform that only hands over a finished transcript can’t provide that detail, so the SDK path captures more than the webhook path.
An eval judge can be given the agent’s system prompt, its tools, its workflow, and the tool calls and results from the call. So an eval can check whether the required tool ran, whether its arguments were right, whether the agent reported what the tool returned, and whether it followed its prompt. Evals can also score tone.
Simulation places voice calls over SIP, inbound or outbound. Your telephony, endpointing, TTS, and latency are all part of the test, so you test what your customers hear.
Managed platforms, self-hosted frameworks, and custom stacks all connect. If you move from a managed platform to your own infrastructure, your evals, history, and dashboards stay in Tuner.
The MCP server reads and writes. You can ask what is failing across your calls, get the root cause, and have the assistant update the agent’s configuration in the same conversation.
Credits are spent on the analyses you switch on. An agent with three evals costs less per call than one with fifteen, and alerts and red flags cost nothing at all. Plans set your monthly credit allowance and feature access. Plans and rates are on the pricing page. See Credits and usage for what uses credits.

Next steps

Quickstart

Connect an agent and get a scored call.

Call Simulation

Test before launch and after every change.