Skip to main content
GET
Get Run

Authorizations

Authorization
string
header
required

Use Tuner API key (tr_api_...) or user session token. Find your API key in Workspace Settings > API Keys.

Path Parameters

run_id
integer
required

ID of the simulation run.

agent_id
string
required

Tuner's numeric agent ID, or your Agent ID from Agent Settings > Agent Connection. Digits-only values are read as the numeric ID.

workspace_id
integer
required

Workspace ID. Find this in Workspace > General Settings.

Response

Successful Response

One simulation run, as in List Runs, plus what a poller needs.

id
integer
required

ID of the simulation run.

status
string
required

Run status: pending, running, analyzing, complete, partial, failed, unreported or analysis_failed. complete means every call finished and was analyzed. running means calls are still executing, or have finished but none has been received yet. analyzing means calls have finished executing and some are still being reported or evaluated. failed, unreported and analysis_failed mean no call was analyzed, because every call failed, no call report arrived, or analysis failed, respectively. failed is also used when scenario generation fails; see failure_reason. partial is a mix of these.

created_at
string<date-time>
required

When the run was created.

failure_reason
string | null
required

Explanation of why the run failed. Null if it did not fail.

functional_count
integer | null
required

Number of functional scenarios requested. Null for runs from saved scenarios.

adversarial_count
integer | null
required

Number of adversarial scenarios requested. Null for runs from saved scenarios.

generated_count
integer
required

Number of test scenarios created so far.

caller_persona_id
integer | null
required

ID of the caller persona used. Null for replay runs.

caller_persona
CallerPersonaResponse · object | null
required

The caller persona used. Null for replay runs.

selected_evals
CriteriaItem · object[]
required

The evals the calls are graded with. Empty if none were selected.

selected_intents
CriteriaItem · object[]
required

The call intents detected on the calls. Empty if none were selected.

passed_count
integer
required

Number of calls that passed their evals.

failed_count
integer
required

Number of calls that failed their evals.

in_progress_count
integer
required

Number of calls still waiting for results (not yet finished or evaluated).

simulation_calls
SimulationCallDetailResponse · object[]
required

The simulated calls in the run, one per scenario.

result
enum<string>
required

Overall result: pending until the run is terminal; then failed if any scenario failed, inconclusive if any scenario could not be judged, else passed.

Available options:
passed,
failed,
inconclusive,
pending
reason
enum<string> | null
required

agent_failed when result is failed, tuner_error when inconclusive, null otherwise.

Available options:
agent_failed,
tuner_error
is_terminal
boolean
required

Whether the run has finished: complete, partial, failed, unreported or analysis_failed. analyzing is not terminal.

dashboard_url
string
required

Link to the run in the Tuner dashboard.

execution_completed_at
string<date-time> | null

When every call finished executing. Null while calls are still in progress. The run can remain running or analyzing after this while calls are reported and evaluated.

simulation_profile_id
integer | null

ID of the simulation profile used, if one was given.

simulation_profile_name
string | null

Name of the simulation profile used, if one was given.

simulation_profile_snapshot
Simulation Profile Snapshot · object | null

Copy of the profile's fields as they were when the run started.

noise_environment_id
integer | null

ID of the background-noise environment used, if one was given.

noise_intensity
number | null

Gain applied to the background noise, if noise was used.

noise_environment
NoiseEnvironmentResponse · object | null

The background-noise environment used, if one was given.

scenario_source
enum<string>
default:generated

generated for generated and saved-scenario runs, recorded_clips for replay runs.

Available options:
generated,
recorded_clips
voice_clip_sets
VoiceClipSetRef · object[]

Voice clip sets played by a replay run. Empty otherwise.

replay_repeat_count
integer | null

Number of times each clip set was replayed. Null for non-replay runs.

replay_settings
Replay Settings · object | null

Turn-timing overrides the replay used. Null for non-replay runs.

evals_expected
boolean
default:true

Whether the calls are graded by evals. False only for replay runs started without evals.

ran_count
integer
default:0

Number of calls that completed in a run started without evals, so there was nothing to grade them.

unreported_count
integer
default:0

Number of calls that completed but whose call data never arrived.

analysis_failed_count
integer
default:0

Number of calls that arrived but whose evals never came back.

run_mode
string
default:generate

generate for runs with generated scenarios, select for library scenarios.