> This is a page from the ElevenLabs documentation. For a complete page index, fetch https://elevenlabs.io/docs/llms.txt. For the full documentation in a single file, fetch https://elevenlabs.io/docs/llms-full.txt.

# Konversationsdetails abrufen

GET https://api.elevenlabs.io/v1/convai/conversations/{conversation_id}

Ruft die Details einer bestimmten Unterhaltung ab

Reference: https://elevenlabs.io/docs/api-reference/conversations/get

## Servers

- `https://api.elevenlabs.io` (Production, default)
- `https://api.us.elevenlabs.io` (Production US)
- `https://api.eu.residency.elevenlabs.io` (Production EU)
- `https://api.in.residency.elevenlabs.io` (Production India)
- `https://api.sg.residency.elevenlabs.io` (Production Singapore)

## Request

### Path parameters

- `conversation_id` (string, required) — Die ID der Unterhaltung, für die Sie die Aktion ausführen.

### Query parameters

- `format` (enum, optional, default: json) — Antwortformat. Standard ist 'json'. Setzen Sie es auf 'opentelemetry' für eine OTLP-kompatible Trace-Payload mit derselben Struktur wie der Webhook nach dem Aufruf.
  - Allowed values: `json`, `opentelemetry`

## Response

### 200

Erfolgreiche Antwort

- `agent_id` (string, required)
- `status` (enum, required)
  - Allowed values: `initiated`, `in-progress`, `processing`, `done`, `failed`
- `metadata` (ConversationHistoryMetadataCommonModel, required)
- `conversation_id` (string, required)
- `has_audio` (boolean, required)
- `has_user_audio` (boolean, required)
- `has_response_audio` (boolean, required)
- `has_auxiliary_audio` (boolean, required)
- `transcript` (list of ConversationHistoryTranscriptResponseModel, required)
- `agent_name` (string, optional, nullable)
- `conversation_product` (string, optional, default: agent)
- `user_id` (string, optional, nullable)
- `branch_id` (string, optional, nullable)
- `version_id` (string, optional, nullable) — Die ID der für diese Unterhaltung verwendeten Agent-Version
- `analysis` (ConversationHistoryAnalysisCommonModel, optional, nullable)
- `visited_agents` (list of VisitedAgentRef, optional)
- `conversation_initiation_client_data` (ConversationInitiationClientDataRequest-Output, optional)
- `environment` (string, optional, default: production)
- `tag_ids` (list of string, optional) — Diesem Gespräch zugewiesene Konversations-Tag-IDs.
- `otlp_traces` (map from string to any, optional, nullable) — OpenTelemetry-Trace-Payload, wenn die Anfrage format=opentelemetry verwendet; andernfalls nicht enthalten.

## Errors

### 422 Unprocessable Entity Error

Validierungsfehler

- `detail` (list of ValidationError, optional)

## Types

### ConversationHistoryMetadataCommonModel

- `start_time_unix_secs` (integer, required)
- `call_duration_secs` (integer, required)
- `cost_fiat` (double, required, nullable) — Gesamte Fiat-Kosten der Unterhaltung in USD, also die Summe aus dem LLM-Preis und dem Nicht-LLM-Plattformpreis (das Fiat-Äquivalent von ``cost``). ``None``, wenn keiner der beiden Werte festgelegt ist, z. B. bei Unterhaltungen vor Einführung der Fiat-Kostenerfassung.
- `accepted_time_unix_secs` (integer, optional, nullable)
- `queue_wait_secs` (double, optional, nullable) — Sekunden, die der Anrufer in der Warteschlange für Gleichzeitigkeit verbrachte. Nicht in call_duration_secs und die abgerechnete Zeit einbezogen. None, wenn das Gespräch nie in die Warteschlange eingereiht wurde.
- `cost` (integer, optional, nullable)
- `deletion_settings` (ConversationDeletionSettings, optional)
- `feedback` (ConversationHistoryFeedbackCommonModel, optional)
- `authorization_method` (enum, optional, default: public)
  - Allowed values: `invalid`, `public`, `authorization_header`, `signed_url`, `shareable_link`, `livekit_token`, `livekit_token_website`, `genesys_api_key`, `avaya_api_key`, `audiocodes_api_key`, `whatsapp`, `sms`
- `charging` (ConversationChargingCommonModel, optional)
- `phone_call` (ConversationHistoryMetadataCommonModelPhoneCall, optional, nullable)
- `batch_call` (ConversationHistoryBatchCallModel, optional, nullable)
- `termination_reason` (string, optional, default: )
- `error` (ConversationHistoryErrorCommonModel, optional, nullable)
- `warnings` (list of string, optional)
- `main_language` (string, optional, nullable)
- `rag_usage` (ConversationHistoryRagUsageCommonModel, optional, nullable)
- `text_only` (boolean, optional, default: false)
- `features_usage` (FeaturesUsageCommonModel, optional)
- `eleven_assistant` (ConversationHistoryElevenAssistantCommonModel, optional)
- `initiator_id` (string, optional, nullable)
- `conversation_initiation_source` (enum, optional, default: unknown) — Aufzählung der möglichen Quellen für den Gesprächsbeginn.
  - Allowed values: `unknown`, `android_sdk`, `node_js_sdk`, `react_native_sdk`, `react_sdk`, `js_sdk`, `python_sdk`, `widget`, `sip_trunk`, `twilio`, `exotel`, `genesys`, `avaya`, `audiocodes`, `swift_sdk`, `whatsapp`, `twilio_sms`, `flutter_sdk`, `zendesk_integration`, `slack_integration`, `telegram_integration`, `intercom_integration`, `freshdesk_integration`, `salesforce_integration`, `template_preview`, `genesys_bot_connector`, `subagent_tool`
- `conversation_initiation_source_version` (string, optional, nullable)
- `timezone` (string, optional, nullable)
- `async_metadata` (AsyncConversationMetadata, optional, nullable) — Metadaten für die asynchrone Zustellung von Unterhaltungen (Zendesk, Slack usw.).
- `whatsapp` (WhatsAppConversationInfo, optional, nullable)
- `sms` (SMSConversationInfo, optional, nullable)
- `agent_created_from` (enum, optional, default: unknown)
  - Allowed values: `cli`, `ui`, `api`, `template`, `unknown`
- `agent_last_updated_from` (enum, optional, default: unknown)
  - Allowed values: `cli`, `ui`, `api`, `template`, `unknown`
- `voice_rewards` (list of ConversationVoiceRewardModel, optional)

### ConversationHistoryTranscriptResponseModel

- `role` (enum, required)
  - Allowed values: `user`, `agent`
- `time_in_call_secs` (integer, required)
- `agent_metadata` (AgentMetadata, optional, nullable)
- `message` (string, optional, nullable)
- `multivoice_message` (ConversationHistoryMultivoiceMessageModel, optional, nullable) — Repräsentiert eine Nachricht eines Multi-Voice-Agenten.
- `tool_calls` (list of ConversationHistoryTranscriptToolCallCommonModel-Output, optional)
- `tool_results` (list of ConversationHistoryTranscriptResponseModelToolResultsItems, optional)
- `feedback` (UserFeedback, optional, nullable)
- `llm_override` (string, optional, nullable)
- `producing_llm` (string, optional, nullable)
- `conversation_turn_metrics` (ConversationTurnMetrics, optional, nullable)
- `rag_retrieval_info` (RagRetrievalInfo, optional, nullable)
- `llm_usage` (LLMUsage-Output, optional, nullable)
- `interrupted` (boolean, optional, default: false)
- `ignored_as_backchannel` (boolean, optional, default: false)
- `platform_event` (enum, optional, nullable)
  - Allowed values: `user_turn_timeout`, `skip_turn_wait_elapsed`
- `original_message` (string, optional, nullable)
- `reasoning` (list of ConversationReasoningModel, optional)
- `source_medium` (enum, optional, nullable)
  - Allowed values: `audio`, `dtmf`, `text`, `image`, `file`
- `source_event_id` (integer, optional, nullable)
- `used_static_kb_document_ids` (list of string, optional)
- `user_identifier` (string, optional, nullable)
- `id` (string, optional, nullable)
- `triggered_guardrails` (list of TriggeredGuardrailCommonModel, optional)
- `file_input` (ConversationHistoryTranscriptFileInputResponseModel, optional, nullable) — Veraltet: der erste Anhang in diesem Turn. Verwenden Sie `file_inputs`, um alle Anhänge anzuzeigen.
- `file_inputs` (list of ConversationHistoryTranscriptFileInputResponseModel, optional) — Alle an diesen Turn angehängten Dateien in der Reihenfolge, in der der Nutzer sie angehängt hat.
- `contextual_update_info` (ContextualUpdateInfo, optional, nullable)
- `reasoned` (boolean, optional, default: false)

### ConversationHistoryAnalysisCommonModel

- `call_successful` (enum, required)
  - Allowed values: `success`, `failure`, `unknown`
- `transcript_summary` (string, required)
- `evaluation_criteria_results` (map from string to ConversationHistoryEvaluationCriteriaResultCommonModel, optional)
- `data_collection_results` (map from string to DataCollectionResultCommonModel, optional)
- `evaluation_criteria_results_list` (list of ConversationHistoryEvaluationCriteriaResultCommonModel, optional)
- `data_collection_results_list` (list of DataCollectionResultCommonModel, optional)
- `call_success_score` (double, optional, nullable)
- `call_summary_title` (string, optional, nullable)
- `scoped` (list of ScopedAnalysisResult, optional)

### VisitedAgentRef

Ein Agent und ein optionaler Branch, die am Anruf teilgenommen haben, in der Reihenfolge ihres ersten Auftretens im Transkript.

- `agent_id` (string, required)
- `branch_id` (string, optional, nullable)

### ConversationInitiationClientDataRequest-Output

- `conversation_config_override` (ConversationConfigClientOverride-Output, optional)
- `custom_llm_extra_body` (map from string to any, optional)
- `user_id` (string, optional, nullable) — ID des Endnutzers, der an dieser Unterhaltung teilnimmt (zur Nutzeridentifikation des Agent-Eigentümers)
- `source_info` (ConversationInitiationSourceInfo, optional) — Informationen über die Quelle der Gesprächsinitiierung
- `branch_id` (string, optional, nullable) — ID des Agent-Branches, der für diese Unterhaltung verwendet werden soll
- `environment` (string, optional, nullable) — Umgebung zum Auflösen von Umgebungsvariablen
- `starting_workflow_node_id` (string, optional, nullable) — Falls gesetzt, startet der Workflow an dieser Knoten-ID statt am Standardeinstieg.
- `procedure_ids` (list of string, optional, nullable) — Wenn gesetzt, stehen dem startenden Agenten nur diese Prozeduren zur Verfügung. Jede ID muss diesem Agenten zugeordnet sein. Unbekannte IDs verhindern den Start der Unterhaltung. Eine leere Liste deaktiviert alle Prozeduren dieses Agenten. Wird nach einer Agentenübertragung nicht angewendet. Erfordert enable_procedure_ids_from_client.
- `dynamic_variables` (map from string to any, optional)

### ValidationError

- `loc` (list of ValidationErrorLocItems, required)
- `msg` (string, required)
- `type` (string, required)

### ConversationDeletionSettings

- `deletion_time_unix_secs` (integer, optional, nullable)
- `deleted_logs_at_time_unix_secs` (integer, optional, nullable)
- `deleted_audio_at_time_unix_secs` (integer, optional, nullable)
- `deleted_transcript_at_time_unix_secs` (integer, optional, nullable)
- `delete_transcript_and_pii` (boolean, optional, default: false)
- `delete_audio` (boolean, optional, default: false)

### ConversationHistoryFeedbackCommonModel

- `type` (enum, optional, nullable)
  - Allowed values: `thumbs`, `rating`
- `overall_score` (enum, optional, nullable)
  - Allowed values: `like`, `dislike`
- `likes` (integer, optional, default: 0)
- `dislikes` (integer, optional, default: 0)
- `rating` (integer, optional, nullable)
- `comment` (string, optional, nullable)

### ConversationChargingCommonModel

- `dev_discount` (boolean, optional, default: false)
- `is_burst` (boolean, optional, default: false)
- `tier` (string, optional, nullable)
- `llm_usage` (LLMCategoryUsage, optional)
- `llm_price` (double, optional, nullable)
- `llm_charge` (integer, optional, nullable)
- `call_charge` (integer, optional, nullable)
- `platform_charge` (integer, optional, nullable)
- `platform_usage` (PlatformUsage, optional) — Aufschlüsselung von ``platform_charge`` nach Kategorie (das Gegenstück zu ``llm_usage``).
- `platform_price` (double, optional, nullable)
- `free_minutes_consumed` (double, optional, default: 0)
- `free_llm_dollars_consumed` (double, optional, default: 0)
- `tts_usage` (ConversationTTSUsageModel, optional, nullable) — Aggregierte TTS-Nutzung für eine Unterhaltung (nur Analysen, keine Abrechnung).
- `asr_usage` (ConversationASRUsageModel, optional, nullable) — Aggregierte ASR-Nutzung für eine Unterhaltung (nur Analysen, keine Abrechnung).
- `analysis` (AnalysisCharging, optional, nullable) — Kosten für die Durchführung einer Nachgesprächsanalyse für dieses Gespräch. Vorhanden, sobald eine Analyse durchgeführt wurde, unabhängig davon, ob sie abgerechnet wurde.

### ConversationHistoryMetadataCommonModelPhoneCall

- `type`: `exotel` (ConversationHistoryExotelPhoneCallModel)
  - `agent_number` (string, required)
  - `call_sid` (string, required)
  - `direction` (enum, required, default: inbound)
    - Allowed values: `inbound`, `outbound`
  - `external_number` (string, required)
  - `phone_number_id` (string, required)
  - `stream_sid` (string, required)
  - `call_result` (enum, optional, nullable)
    - Allowed values: `answered`, `voicemail`, `busy`, `no_answer`, `failed`
- `type`: `sip_trunking` (ConversationHistorySIPTrunkingPhoneCallModel)
  - `agent_number` (string, required)
  - `call_sid` (string, required)
  - `direction` (enum, required, default: inbound)
    - Allowed values: `inbound`, `outbound`
  - `external_number` (string, required)
  - `phone_number_id` (string, required)
  - `call_id` (string, optional, nullable)
  - `call_result` (enum, optional, nullable)
    - Allowed values: `answered`, `voicemail`, `busy`, `no_answer`, `failed`
  - `sip_header_dynamic_variables` (map from string to string, optional)
- `type`: `twilio` (ConversationHistoryTwilioPhoneCallModel)
  - `agent_number` (string, required)
  - `call_sid` (string, required)
  - `direction` (enum, required, default: inbound)
    - Allowed values: `inbound`, `outbound`
  - `external_number` (string, required)
  - `phone_number_id` (string, required)
  - `stream_sid` (string, required)
  - `call_result` (enum, optional, nullable)
    - Allowed values: `answered`, `voicemail`, `busy`, `no_answer`, `failed`

### ConversationHistoryBatchCallModel

- `batch_call_id` (string, required)
- `batch_call_recipient_id` (string, required)
- `campaign` (BatchCallingCampaignInformation, optional, nullable)

### ConversationHistoryErrorCommonModel

- `code` (integer, required)
- `reason` (string, optional, nullable)

### ConversationHistoryRagUsageCommonModel

- `usage_count` (integer, required)
- `embedding_model` (string, required)

### FeaturesUsageCommonModel

- `language_detection` (FeatureStatusCommonModel, optional)
- `transfer_to_agent` (FeatureStatusCommonModel, optional)
- `transfer_to_number` (FeatureStatusCommonModel, optional)
- `multivoice` (FeatureStatusCommonModel, optional)
- `dtmf_tones` (FeatureStatusCommonModel, optional)
- `external_mcp_servers` (FeatureStatusCommonModel, optional)
- `pii_zrm_workspace` (boolean, optional, default: false)
- `pii_zrm_agent` (boolean, optional, default: false)
- `tool_dynamic_variable_updates` (FeatureStatusCommonModel, optional)
- `is_livekit` (boolean, optional, default: false)
- `voicemail_detection` (FeatureStatusCommonModel, optional)
- `dtmf_input` (FeatureStatusCommonModel, optional)
- `workflow` (WorkflowFeaturesUsageCommonModel, optional)
- `agent_testing` (TestsFeatureUsageCommonModel, optional)
- `versioning` (FeatureStatusCommonModel, optional)
- `file_input` (FeatureStatusCommonModel, optional)
- `freeform_procedure` (FeatureStatusCommonModel, optional)
- `structured_procedure` (FeatureStatusCommonModel, optional)

### ConversationHistoryElevenAssistantCommonModel

- `is_eleven_assistant` (boolean, optional, default: false)

### AsyncConversationMetadata

Metadaten für die asynchrone Zustellung von Unterhaltungen (Zendesk, Slack usw.).

- `delivery_status` (enum, required)
  - Allowed values: `pending`, `success`, `failed`
- `delivery_timestamp` (integer, required)
- `external_system` (string, required)
- `external_id` (string, required)
- `delivery_error` (string, optional, nullable)
- `external_link` (string, optional, nullable)
- `retry_count` (integer, optional, default: 0)
- `last_retry_timestamp` (integer, optional, nullable)
- `last_processed_external_message_id` (string, optional, nullable)

### WhatsAppConversationInfo

- `whatsapp_user_id` (string, required)
- `direction` (enum, optional, default: unknown)
  - Allowed values: `inbound`, `outbound`, `unknown`
- `whatsapp_phone_number_id` (string, optional, nullable)
- `awaiting_first_user_message` (boolean, optional, nullable)
- `recipient` (WhatsAppRecipient, optional, nullable) — Ein WhatsApp-Nutzer per Telefonnummer und/oder BSUID; wenn nur eines erforderlich ist, hat die Telefonnummer Vorrang.
- `call_permission_request` (WhatsAppCallPermissionRequest, optional, nullable)

### SMSConversationInfo

- `direction` (enum, required)
  - Allowed values: `inbound`, `outbound`
- `sms_user_phone_number` (string, required)
- `phone_number_id` (string, optional, nullable)
- `agent_phone_number` (string, optional, nullable)

### ConversationVoiceRewardModel

- `voice_id` (string, required)
- `reward_usd_cents` (double, required)

### AgentMetadata

- `agent_id` (string, required)
- `branch_id` (string, optional, nullable)
- `workflow_node_id` (string, optional, nullable)
- `version_id` (string, optional, nullable)

### ConversationHistoryMultivoiceMessageModel

Repräsentiert eine Nachricht eines Multi-Voice-Agenten.

- `parts` (list of ConversationHistoryMultivoiceMessagePartModel, required)

### ConversationHistoryTranscriptToolCallCommonModel-Output

- `request_id` (string, required)
- `tool_name` (string, required)
- `params_as_json` (string, required)
- `tool_has_been_called` (boolean, required)
- `type` (enum, optional, nullable)
  - Allowed values: `system`, `webhook`, `client`, `mcp`, `workflow`, `api_integration_webhook`, `api_integration_mcp`, `smb`
- `tool_id` (string, optional, nullable) — ID des aufgerufenen Tools. Für Tools ohne ID, etwa Systemtools, nicht festgelegt.
- `tool_details` (ConversationHistoryTranscriptToolCallCommonModelOutputToolDetails, optional, nullable)

### ConversationHistoryTranscriptResponseModelToolResultsItems

### UserFeedback

- `score` (enum, required)
  - Allowed values: `like`, `dislike`
- `time_in_call_secs` (integer, required)

### ConversationTurnMetrics

- `metrics` (map from string to MetricRecord, optional)
- `convai_asr_provider` (string, optional, nullable)
- `convai_tts_model` (string, optional, nullable)
- `convai_tts_cascade` (string, optional, nullable)
- `voice_id` (string, optional, nullable)

### RagRetrievalInfo

- `chunks` (list of RagChunkMetadata, required)
- `embedding_model` (enum, required, default: e5_mistral_7b_instruct)
  - Allowed values: `e5_mistral_7b_instruct`, `multilingual_e5_large_instruct`
- `retrieval_query` (string, required)
- `rag_latency_secs` (double, required)
- `used_chunk_ids` (list of string, optional)

### LLMUsage-Output

- `model_usage` (map from string to LLMInputOutputTokensUsage, optional)

### ConversationReasoningModel

- `summary` (string, optional, nullable)
- `provider_redact` (boolean, optional, default: false)

### TriggeredGuardrailCommonModel

- `guardrail_type` (enum, required)
  - Allowed values: `custom`, `prompt_injection`, `self_harm_intent`, `violence_graphic`, `sexual`, `violence`, `harassment`, `sexual_minors`, `self_harm`, `self_harm_instructions`, `harassment_threatening`, `hate`, `hate_threatening`, `profanity`, `religion_or_politics`, `medical_and_legal`, `guardrail`
- `guardrail_name` (string, optional, nullable)

### ConversationHistoryTranscriptFileInputResponseModel

- `file_id` (string, required)
- `original_filename` (string, required)
- `mime_type` (string, required)
- `file_url` (string, required)

### ContextualUpdateInfo

- `context_id` (string, required) — Vom Client bereitgestellter Identifier zur Gruppierung zusammengehöriger Kontextaktualisierungen.
- `is_superseded` (boolean, optional, default: false) — True, wenn diese kontextbezogene Aktualisierung durch eine neuere Aktualisierung mit derselben context_id ersetzt wurde.

### ConversationHistoryEvaluationCriteriaResultCommonModel

- `criteria_id` (string, required)
- `result` (enum, required)
  - Allowed values: `success`, `failure`, `unknown`
- `rationale` (string, required)
- `scoring_mode` (enum, optional, nullable, default: binary)
  - Allowed values: `binary`, `numeric_uniform`
- `score` (integer, optional, nullable)
- `max_score` (integer, optional, nullable)

### DataCollectionResultCommonModel

- `data_collection_id` (string, required)
- `rationale` (string, required)
- `name` (string, optional, nullable)
- `value` (any, optional)
- `json_schema` (LiteralJsonSchemaProperty, optional, nullable) — Schema-Eigenschaft für literale JSON-Typen. WICHTIG: Es kann nur EINES der folgenden Felder festgelegt werden: description (LLM liefert Wert), dynamic_variable (Wert aus Variable), is_system_provided (System liefert Wert), constant_value (fester Wert) oder is_omitted (Parameter wird ausgelassen). Diese schließen sich gegenseitig aus.

### ScopedAnalysisResult

- `scope` (enum, required, default: conversation) — Der Umfang der Analyse. 'conversation' verwendet das vollständige Transkript; 'agent' verwendet nur den Abschnitt, in dem der definierende Agent aktiv war.
  - Allowed values: `conversation`, `agent`
- `source_agent_id` (string, required)
- `successful` (enum, required)
  - Allowed values: `success`, `failure`, `unknown`
- `source_branch_id` (string, optional, nullable) — Branch des Agents für diesen Bereich; unterscheidet wiederholte agent_id.
- `evaluation_criteria_results` (map from string to ConversationHistoryEvaluationCriteriaResultCommonModel, optional)
- `data_collection_results` (map from string to DataCollectionResultCommonModel, optional)
- `success_score` (double, optional, nullable)

### ConversationConfigClientOverride-Output

- `asr` (ASRConversationalConfigOverride, optional, nullable) — Konfiguration für Transkription in Unterhaltungen
- `turn` (TurnConfigOverride, optional, nullable) — Konfiguration für Sprecherwechselerkennung
- `tts` (TTSConversationalConfigOverride, optional, nullable) — Konfiguration für Text to Speech in Unterhaltungen
- `conversation` (ConversationConfigOverride, optional, nullable) — Konfiguration für Unterhaltungsereignisse
- `agent` (AgentConfigOverride-Output, optional, nullable) — Agent-spezifische Konfiguration

### ConversationInitiationSourceInfo

Informationen über die Quelle der Gesprächsinitiierung

- `source` (enum, optional, nullable, default: unknown) — Quelle der Gesprächsinitiierung
  - Allowed values: `unknown`, `android_sdk`, `node_js_sdk`, `react_native_sdk`, `react_sdk`, `js_sdk`, `python_sdk`, `widget`, `sip_trunk`, `twilio`, `exotel`, `genesys`, `avaya`, `audiocodes`, `swift_sdk`, `whatsapp`, `twilio_sms`, `flutter_sdk`, `zendesk_integration`, `slack_integration`, `telegram_integration`, `intercom_integration`, `freshdesk_integration`, `salesforce_integration`, `template_preview`, `genesys_bot_connector`, `subagent_tool`
- `version` (string, optional, nullable) — Die SDK-Versionsnummer

### ValidationErrorLocItems

### LLMCategoryUsage

- `irreversible_generation` (LLMUsage-Output, optional)
- `initiated_generation` (LLMUsage-Output, optional)

### PlatformUsage

Aufschlüsselung von ``platform_charge`` nach Kategorie (das Gegenstück zu ``llm_usage``).

- `category_usage` (map from string to PlatformCategoryUsage, optional)

### ConversationTTSUsageModel

Aggregierte TTS-Nutzung für eine Unterhaltung (nur Analysen, keine Abrechnung).

- `primary_tts_model` (string, optional, nullable)
- `total_audio_output_seconds` (double, optional, default: 0)
- `total_characters` (integer, optional, default: 0)
- `per_voice_usage` (list of ConversationVoiceUsageModel, optional)

### ConversationASRUsageModel

Aggregierte ASR-Nutzung für eine Unterhaltung (nur Analysen, keine Abrechnung).

- `asr_model` (string, optional, nullable)
- `total_transcription_calls` (integer, optional, default: 0)
- `total_audio_input_seconds` (double, optional, default: 0)

### AnalysisCharging

Kosten für die Durchführung einer Nachgesprächsanalyse für dieses Gespräch. Vorhanden, sobald eine Analyse durchgeführt wurde, unabhängig davon, ob sie abgerechnet wurde.

- `total` (AnalysisRunningTotal, required) — Kumulierte LLM-Kosten für die Durchführung der Nachbearbeitungsanalyse dieses Gesprächs.
- `last_run` (AnalysisRunSnapshot, required) — LLM-Kosten der letzten Analyse dieser Konversation nach dem Anruf.

### BatchCallingCampaignInformation

- `campaign_id` (string, required)
- `campaign_lead_id` (string, required)
- `cold_attempt` (integer, optional, nullable)
- `conversation_type` (enum, optional, nullable)
  - Allowed values: `initial`, `retry`, `callback`, `redirect`

### FeatureStatusCommonModel

- `enabled` (boolean, optional, default: false)
- `used` (boolean, optional, default: false)

### WorkflowFeaturesUsageCommonModel

- `enabled` (boolean, optional, default: false)
- `tool_node` (FeatureStatusCommonModel, optional)
- `standalone_agent_node` (FeatureStatusCommonModel, optional)
- `phone_number_node` (FeatureStatusCommonModel, optional)
- `end_node` (FeatureStatusCommonModel, optional)
- `override_agent_node` (FeatureStatusCommonModel, optional)

### TestsFeatureUsageCommonModel

- `enabled` (boolean, optional, default: false)
- `tests_ran_after_last_modification` (boolean, optional, default: false)
- `tests_ran_in_last_7_days` (boolean, optional, default: false)

### WhatsAppRecipient

Ein WhatsApp-Nutzer per Telefonnummer und/oder BSUID; wenn nur eines erforderlich ist, hat die Telefonnummer Vorrang.

- `phone` (string, optional, nullable)
- `bsuid` (string, optional, nullable)

### WhatsAppCallPermissionRequest

- `state` (enum, required)
  - Allowed values: `blocked`, `sent`, `accepted`, `declined`
- `requested_at_unix` (integer, optional, nullable)

### ConversationHistoryMultivoiceMessagePartModel

Repräsentiert einen einzelnen Stimmanteil einer Multi-Voice-Nachricht.

- `text` (string, required)
- `voice_label` (string, required, nullable)
- `time_in_call_secs` (integer, required, nullable)

### ConversationHistoryTranscriptToolCallCommonModelOutputToolDetails

- `type`: `api_integration_webhook` (ConversationHistoryTranscriptToolCallApiIntegrationWebhookDetails)
  - `credential_id` (string, required, default: )
  - `integration_connection_id` (string, required, default: )
  - `integration_id` (string, required, default: )
  - `webhook_details` (ConversationHistoryTranscriptToolCallWebhookDetails, required)
- `type`: `client` (ConversationHistoryTranscriptToolCallClientDetails)
  - `parameters` (string, required)
- `type`: `mcp` (ConversationHistoryTranscriptToolCallMCPDetails)
  - `approval_policy` (string, required)
  - `integration_type` (string, required)
  - `mcp_server_id` (string, required)
  - `mcp_server_name` (string, required)
  - `mcp_tool_description` (string, optional, default: )
  - `mcp_tool_name` (string, optional, default: )
  - `parameters` (map from string to string, optional)
  - `requires_approval` (boolean, optional, default: false)
- `type`: `webhook` (ConversationHistoryTranscriptToolCallWebhookDetails)
  - `method` (string, required)
  - `url` (string, required)
  - `body` (string, optional, nullable)
  - `headers` (map from string to string, optional)
  - `path_params` (map from string to string, optional)
  - `query_params` (map from string to string, optional)

### ConversationHistoryTranscriptOtherToolsResultCommonModel

- `request_id` (string, required)
- `tool_name` (string, required)
- `result_value` (string, required)
- `is_error` (boolean, required)
- `tool_has_been_called` (boolean, required)
- `is_blocked` (boolean, optional, default: false)
- `tool_latency_secs` (double, optional, default: 0)
- `error_type` (string, optional, default: )
- `raw_error_message` (string, optional, default: )
- `dynamic_variable_updates` (list of DynamicVariableUpdateCommonModel, optional)
- `type` (enum, optional, nullable)
  - Allowed values: `client`, `webhook`, `mcp`, `code`

### ConversationHistoryTranscriptSystemToolResultCommonModel-Output

- `request_id` (string, required)
- `tool_name` (string, required)
- `result_value` (string, required)
- `is_error` (boolean, required)
- `tool_has_been_called` (boolean, required)
- `type` ("system", required)
- `is_blocked` (boolean, optional, default: false)
- `tool_latency_secs` (double, optional, default: 0)
- `error_type` (string, optional, default: )
- `raw_error_message` (string, optional, default: )
- `dynamic_variable_updates` (list of DynamicVariableUpdateCommonModel, optional)
- `result` (ConversationHistoryTranscriptSystemToolResultCommonModelOutputResult, optional, nullable)

### ConversationHistoryTranscriptApiIntegrationWebhookToolsResultCommonModel-Output

- `request_id` (string, required)
- `tool_name` (string, required)
- `result_value` (string, required)
- `is_error` (boolean, required)
- `is_blocked` (boolean, required, default: false)
- `tool_has_been_called` (boolean, required)
- `tool_latency_secs` (double, required, default: 0)
- `error_type` (string, required, default: )
- `raw_error_message` (string, required, default: )
- `dynamic_variable_updates` (list of DynamicVariableUpdateCommonModel, required)
- `type` ("api_integration_webhook", required)
- `integration_id` (string, required, default: )
- `credential_id` (string, required, default: )
- `integration_connection_id` (string, required, default: )

### ConversationHistoryTranscriptWorkflowToolsResultCommonModel-Output

- `request_id` (string, required)
- `tool_name` (string, required)
- `result_value` (string, required)
- `is_error` (boolean, required)
- `tool_has_been_called` (boolean, required)
- `type` ("workflow", required)
- `is_blocked` (boolean, optional, default: false)
- `tool_latency_secs` (double, optional, default: 0)
- `error_type` (string, optional, default: )
- `raw_error_message` (string, optional, default: )
- `dynamic_variable_updates` (list of DynamicVariableUpdateCommonModel, optional)
- `result` (WorkflowToolResponseModel-Output, optional, nullable) — Ein gemeinsames Modell für Antworten von Workflow-Tools.

### MetricRecord

- `elapsed_time` (double, required)

### RagChunkMetadata

- `document_id` (string, required)
- `chunk_id` (string, required)
- `vector_distance` (double, required)

### LLMInputOutputTokensUsage

- `input` (LLMTokensCategoryUsage, optional)
- `input_cache_read` (LLMTokensCategoryUsage, optional)
- `input_cache_write` (LLMTokensCategoryUsage, optional)
- `output_total` (LLMTokensCategoryUsage, optional)

### LiteralJsonSchemaProperty

Schema-Eigenschaft für literale JSON-Typen. WICHTIG: Es kann nur EINES der folgenden Felder festgelegt werden: description (LLM liefert Wert), dynamic_variable (Wert aus Variable), is_system_provided (System liefert Wert), constant_value (fester Wert) oder is_omitted (Parameter wird ausgelassen). Diese schließen sich gegenseitig aus.

- `type` (LiteralJsonSchemaPropertyType, required)
- `description` (string, optional, default: ) — Die Beschreibung der Eigenschaft. Wenn festgelegt, stellt das LLM den Wert anhand dieser Beschreibung bereit. Nicht gleichzeitig mit dynamic_variable, is_system_provided, constant_value und is_omitted verwendbar.
- `enum` (list of string, optional, nullable) — List of allowed string values for string type parameters
- `is_system_provided` (boolean, optional, default: false) — Wenn true, wird der Wert vom System zur Laufzeit ausgefüllt. Wird von API-Integration-Webhook-Tools für Vorlagen verwendet. Schließt sich gegenseitig mit description, dynamic_variable, constant_value und is_omitted aus.
- `dynamic_variable` (string, optional, default: ) — Der Name der dynamischen Variable, die für den Wert dieser Eigenschaft verwendet werden soll. Kann nicht zusammen mit description, is_system_provided, constant_value und is_omitted verwendet werden.
- `allowed_values` (AllowedValues, optional, nullable) — Serverseitige Ablehnungsprüfung für einen von der LLM bereitgestellten Wert: Die Laufzeit lehnt jeden Wert außerhalb der von diesem Objekt benannten zulässigen Menge ab. Die Menge wird der LLM nicht als Enum mitgeteilt. Wird nur unterstützt, wenn die Wertquelle `description` ist; eine Kombination mit dynamic_variable, is_system_provided, constant_value oder is_omitted wird abgelehnt.
- `constant_value` (LiteralJsonSchemaPropertyConstantValue, optional, nullable, default: ) — Ein konstanter Wert für diese Eigenschaft. Schließt sich gegenseitig mit description, dynamic_variable, is_system_provided und is_omitted aus.
- `is_omitted` (boolean, optional, default: false) — Wenn true, wird dieser Parameter vollständig aus der Anfrage weggelassen. Nur für optionale Parameter gültig. Schließt sich gegenseitig mit description, dynamic_variable, is_system_provided und constant_value aus.
- `allowed_values_dynamic_variable` (string, optional, default: , deprecated) — VERALTET: Verwenden Sie stattdessen `allowed_values`. Wenn festgelegt, liefert das LLM den Wert, aber die Laufzeitumgebung lehnt jeden Wert ab, der nicht in der Liste dieser dynamischen Variable enthalten ist (muss ein JSON-Array wie ["ws_alpha", "ws_beta"] sein). Verwenden Sie dies, damit das LLM aus einer serverseitig verifizierten Menge auswählen kann (z. B. den IDs, auf die der aktuelle Benutzer zugreifen darf). Erfordert description; schließt sich gegenseitig mit dynamic_variable, is_system_provided, constant_value und is_omitted aus.

### ASRConversationalConfigOverride

- `keywords` (list of string, optional, nullable) — Schlüsselwörter, deren Vorhersagewahrscheinlichkeit erhöht werden soll

### TurnConfigOverride

- `soft_timeout_config` (SoftTimeoutConfigOverride, optional, nullable) — Konfiguration für die Soft-Timeout-Funktion. Liefert sofortiges Feedback bei längeren LLM-Antworten.

### TTSConversationalConfigOverride

- `model_id` (enum, optional, nullable, default: eleven_v4_turbo) — Das für TTS zu verwendende Modell
  - Allowed values: `eleven_turbo_v2`, `eleven_turbo_v2_5`, `eleven_flash_v2`, `eleven_flash_v2_5`, `eleven_multilingual_v2`, `eleven_v3_conversational`, `eleven_v4`, `eleven_v4_turbo`
- `voice_id` (string, optional, nullable) — Die für TTS zu verwendende Stimmen-ID
- `supported_voices` (list of SupportedVoice, optional, nullable) — Zusätzlich unterstützte Stimmen für den Agenten
- `stability` (double, optional, nullable) — Die Stabilität der generierten Sprache
- `speed` (double, optional, nullable) — Die Geschwindigkeit der generierten Sprache
- `similarity_boost` (double, optional, nullable) — Der Ähnlichkeits-Boost für generierte Sprache
- `pronunciation_dictionary_locators` (list of PydanticPronunciationDictionaryVersionLocator, optional, nullable) — Die Referenzen auf Aussprachewörterbücher

### ConversationConfigOverride

- `text_only` (boolean, optional, nullable) — Wenn aktiviert, wird Audio nicht verarbeitet und nur Text verwendet. Verwenden Sie dies, um Audiokosten zu vermeiden.
- `max_duration_seconds` (integer, optional, nullable) — Die maximale Dauer einer Unterhaltung in Sekunden

### AgentConfigOverride-Output

- `first_message` (string, optional, nullable) — Wenn nicht leer, die erste Nachricht, die der Agent sagt. Wenn leer, wartet der Agent, bis der Nutzer die Unterhaltung beginnt.
- `language` (string, optional, nullable) — Sprache des Agenten – verwendet für ASR und TTS
- `max_conversation_duration_message` (string, optional, nullable) — Wenn nicht leer, die Nachricht, die der Agent sendet, wenn die maximale Gesprächsdauer erreicht ist.
- `prompt` (PromptAgentAPIModelOverride-Output, optional, nullable) — Der Prompt für den Agenten

### PlatformCategoryUsage

Aufgelaufene Gebühr für eine einzelne :class:`PlatformCategory`.

- `credits` (integer, optional, default: 0)
- `price` (double, optional, default: 0)
- `quantity` (double, optional, default: 0)

### ConversationVoiceUsageModel

- `voice_id` (string, required)
- `audio_output_seconds` (double, optional, default: 0)

### AnalysisRunningTotal

Kumulierte LLM-Kosten für die Durchführung der Nachbearbeitungsanalyse dieses Gesprächs.

- `price` (double, optional, default: 0)
- `charge` (integer, optional, default: 0)
- `runs` (integer, optional, default: 0)
- `price_per_feature` (map from string to double, optional)
- `charge_per_feature` (map from string to integer, optional)

### AnalysisRunSnapshot

LLM-Kosten der letzten Analyse dieser Konversation nach dem Anruf.

- `price` (double, optional, default: 0)
- `charge` (integer, optional, default: 0)
- `price_per_feature` (map from string to double, optional)
- `charge_per_feature` (map from string to integer, optional)

### ConversationHistoryTranscriptToolCallWebhookDetails

- `method` (string, required)
- `url` (string, required)
- `headers` (map from string to string, optional)
- `path_params` (map from string to string, optional)
- `query_params` (map from string to string, optional)
- `body` (string, optional, nullable)

### DynamicVariableUpdateCommonModel

Erfasst eine Aktualisierung einer dynamischen Variablen während der Tool-Ausführung.

- `variable_name` (string, required)
- `old_value` (string, required, nullable)
- `new_value` (string, required)
- `updated_at` (double, required)
- `tool_name` (string, required)
- `tool_request_id` (string, required)

### ConversationHistoryTranscriptSystemToolResultCommonModelOutputResult

- `result_type`: `dummy` (DummyToolResultModel)
- `result_type`: `end_call_success` (EndCallToolResultModel)
  - `message` (string, optional, nullable)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `end_procedure_error` (EndProcedureToolResultErrorModel)
  - `message` (string, required)
  - `status` (enum, required)
    - Allowed values: `not_found`, `invalid_id`
  - `procedure_id` (string, optional, nullable)
- `result_type`: `end_procedure_success` (EndProcedureToolResultSuccessModel)
  - `procedure_id` (string, required)
  - `procedure_name` (string, required)
  - `message` (string, optional, default: )
  - `status` ("success", optional, default: success)
- `result_type`: `knowledge_base_rag_success` (KnowledgeBaseRagToolResultModel)
  - `chunk_count` (integer, optional, default: 0) — Anzahl abgerufener relevanter Chunks
  - `chunks` (list of KnowledgeBaseRagChunkModel, optional) — Abgerufene Chunks; nur im Modus rag-result-in-tool-result befüllt
  - `message` (string, optional, default: Referenced knowledge base.) — Für das LLM lesbarer Status zu den Suchergebnissen
  - `status` (enum, optional, default: success)
    - Allowed values: `success`, `no_documents`, `no_results`
- `result_type`: `knowledge_base_success` (KnowledgeBaseToolResultModel)
  - `chunk_count` (integer, optional, default: 0)
  - `message` (string, optional, default: Referenced knowledge base.)
  - `status` (enum, optional, default: success)
    - Allowed values: `success`, `no_matching_documents`, `no_results`
- `result_type`: `language_detection_success` (LanguageDetectionToolResultModel)
  - `language` (string, optional, nullable)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `play_dtmf_error` (PlayDTMFResultErrorModel)
  - `error` (string, required)
  - `details` (string, optional, nullable)
  - `status` ("error", optional, default: error)
- `result_type`: `play_dtmf_success` (PlayDTMFResultSuccessModel)
  - `dtmf_tones` (string, required)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `skip_turn_success` (SkipTurnToolResponseModel)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `start_procedure_error` (StartProcedureToolResultErrorModel)
  - `message` (string, required)
  - `status` (enum, required)
    - Allowed values: `not_found`, `invalid_name`, `already_active`
  - `procedure_id` (string, optional, nullable)
- `result_type`: `start_procedure_success` (StartProcedureToolResultSuccessModel)
  - `procedure_id` (string, required)
  - `procedure_name` (string, required)
  - `message` (string, optional, default: Procedure is now started. Follow its instructions from the <active-procedures> section of your system prompt.)
  - `procedure_entry_workflow_node` (string, optional, nullable)
  - `procedure_return_workflow_node` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `testing_tool_result` (TestToolResultModel)
  - `reason` (string, optional, default: Skipping tool call in test mode)
  - `status` ("success", optional, default: success)
- `result_type`: `transfer_to_agent_error` (TransferToAgentToolResultErrorModel)
  - `error` (string, required)
  - `from_agent` (string, required)
  - `status` ("error", optional, default: error)
- `result_type`: `transfer_to_agent_success` (TransferToAgentToolResultSuccessModel)
  - `condition` (string, required)
  - `from_agent` (string, required)
  - `to_agent` (string, required)
  - `branch_info` (TransferToAgentToolResultSuccessModelOutputBranchInfo, optional, nullable)
  - `delay_ms` (integer, optional, default: 0)
  - `enable_transferred_agent_first_message` (boolean, optional, default: false)
  - `preserve_client_tts_overrides` (boolean, optional, default: false)
  - `preserve_voice_settings` (boolean, optional, default: false)
  - `status` ("success", optional, default: success)
  - `to_node` (string, optional, nullable)
  - `transfer_message` (string, optional, nullable)
- `result_type`: `transfer_to_number_amazon_connect_success` (TransferToNumberResultAmazonConnectSuccessModel)
  - `client_message` (string, optional, nullable) — Nachricht, die dem Anrufer vorgelesen wird, bevor die Sitzung an Amazon Connect zurückgegeben wird.
  - `note` (string, optional, nullable) — Wird gesetzt, wenn der Agent eine ungültige Option wiederholt hat und die einzige konfigurierte Option verwendet wurde.
  - `reason` (string, optional, nullable) — Grund für die Eskalation durch den Agenten, der mit dem Ergebnis Escalate an Amazon Connect gesendet wird.
  - `status` ("success", optional, default: success)
- `result_type`: `transfer_to_number_error` (TransferToNumberResultErrorModel)
  - `error` (string, required)
  - `details` (string, optional, nullable)
  - `status` ("error", optional, default: error)
- `result_type`: `transfer_to_number_exotel_success` (TransferToNumberResultExotelSuccessModel)
  - `transfer_number` (string, required)
  - `agent_message` (string, optional, nullable)
  - `note` (string, optional, nullable)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `transfer_to_number_sip_success` (TransferToNumberResultSipSuccessModel)
  - `transfer_number` (string, required)
  - `note` (string, optional, nullable)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `transfer_to_number_twilio_success` (TransferToNumberResultTwilioSuccessModel)
  - `agent_message` (string, required)
  - `conference_name` (string, required)
  - `transfer_number` (string, required)
  - `client_message` (string, optional, nullable)
  - `note` (string, optional, nullable)
  - `post_dial_digits` (string, optional, nullable)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
- `result_type`: `voicemail_detection_success` (VoiceMailDetectionResultSuccessModel)
  - `reason` (string, optional, nullable)
  - `status` ("success", optional, default: success)
  - `voicemail_message` (string, optional, nullable)

### WorkflowToolResponseModel-Output

Ein gemeinsames Modell für Antworten von Workflow-Tools.

- `steps` (list of WorkflowToolResponseModelOutputStepsItems, optional)

### LLMTokensCategoryUsage

- `tokens` (integer, optional, default: 0)
- `price` (double, optional, default: 0)

### LiteralJsonSchemaPropertyType

### AllowedValues

- `dynamic_variable` (string, required) — Name einer dynamischen Variablen, die zu einem JSON-Array zulässiger Werte aufgelöst werden muss, z. B. ["ws_alpha", "ws_beta"]. Systemvariablen funktionieren nur, wenn sie zu einer Liste aufgelöst werden.

### LiteralJsonSchemaPropertyConstantValue

Ein konstanter Wert für diese Eigenschaft. Schließt sich gegenseitig mit description, dynamic_variable, is_system_provided und is_omitted aus.

### SoftTimeoutConfigOverride

- `message` (string, optional, nullable) — Nachricht, die angezeigt wird, wenn beim Warten auf die LLM-Antwort der erste weiche Timeout erreicht wird. Unterstützt dynamische Variablen (z. B. \{\{system\_\_time}}, \{\{custom\_variable}}).
- `additional_soft_timeout_messages` (list of string, optional, nullable) — Zusätzliche statische Füllnachrichten für nachfolgende Soft-Timeouts in derselben LLM-Generierung. Das erste Timeout verwendet `message`. Wenn weniger Nachrichten als `max_soft_timeouts_per_generation` konfiguriert sind, wird die zuletzt konfigurierte Nachricht wiederholt; andernfalls wird eine integrierte Füllnachricht verwendet.

### SupportedVoice

- `label` (string, required)
- `voice_id` (string, required)
- `description` (string, optional, nullable)
- `language` (string, optional, nullable)
- `model_family` (enum, optional, nullable)
  - Allowed values: `turbo`, `flash`, `multilingual`, `v3_conversational`, `v4`, `v4_turbo`
- `optimize_streaming_latency` (enum, optional, nullable)
  - Allowed values: `0`, `1`, `2`, `3`, `4`
- `stability` (double, optional, nullable)
- `speed` (double, optional, nullable)
- `similarity_boost` (double, optional, nullable)

### PydanticPronunciationDictionaryVersionLocator

Eine Referenz, damit andere Dokumente auf ein bestimmtes Wörterbuch und dessen Version verweisen können. Dies ist eine Pydantic-Version von PronunciationDictionaryVersionLocatorDBModel. Erforderlich, um die Kompatibilität mit den übrigen Agent-Datenmodellen sicherzustellen.

- `pronunciation_dictionary_id` (string, required) — Die ID des Aussprachewörterbuchs
- `version_id` (string, required, nullable) — Die ID der Version des Aussprachewörterbuchs

### PromptAgentAPIModelOverride-Output

- `prompt` (string, optional, nullable) — Der Prompt für den Agenten
- `llm` (enum, optional, nullable) — Das LLM, das mit dem Prompt und dem Chatverlauf abgefragt wird. Bei Verwendung von Datenresidenz muss das LLM in der Datenresidenzumgebung unterstützt werden.
  - Allowed values: `gpt-4o-mini`, `gpt-4o`, `gpt-4`, `gpt-4-turbo`, `gpt-4.1`, `gpt-4.1-mini`, `gpt-4.1-nano`, `gpt-5`, `gpt-5.1`, `gpt-5.2`, `gpt-5.2-chat-latest`, `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.5`, `gpt-5.6-sol`, `gpt-5.6-terra`, `gpt-5.6-luna`, `gpt-6-astra`, `gpt-6-sol`, `gpt-6-luna`, `gpt-6.1-sol`, `gpt-5-mini`, `gpt-5-nano`, `gpt-3.5-turbo`, `gemini-1.5-pro`, `gemini-1.5-flash`, `gemini-2.0-flash`, `gemini-2.0-flash-lite`, `gemini-2.5-flash-lite`, `gemini-2.5-flash`, `gemini-3-pro-preview`, `gemini-3-flash-preview`, `gemini-3.1-pro-preview`, `gemini-3.1-flash-lite-preview`, `gemini-3.1-flash-lite`, `gemini-3.5-flash`, `gemini-3.5-flash-lite`, `gemini-3.6-flash`, `gemini-3.7-flash`, `gemini-3.8-flash`, `gemini-4-argon`, `claude-sonnet-4-5`, `claude-opus-4-6`, `claude-opus-4-7`, `claude-opus-4-8`, `claude-opus-5`, `claude-opus-5-5`, `claude-sonnet-4-6`, `claude-sonnet-5`, `claude-sonnet-5-5`, `claude-sonnet-4`, `claude-haiku-4-5`, `claude-haiku-5-5`, `claude-3-7-sonnet`, `claude-3-5-sonnet`, `claude-3-5-sonnet-v1`, `claude-3-haiku`, `grok-beta`, `custom-llm`, `qwen3-4b`, `qwen3-30b-a3b`, `qwen36-35b-a3b`, `qwen35-397b-a17b`, `gpt-oss-20b`, `gpt-oss-120b`, `glm-45-air-fp8`, `glm-52`, `deepseek-v41-flash`, `nemotron-35-super-vl-nvfp4`, `gemini-2.5-flash-preview-09-2025`, `gemini-2.5-flash-lite-preview-09-2025`, `gemini-2.5-flash-preview-05-20`, `gemini-2.5-flash-preview-04-17`, `gemini-2.5-flash-lite-preview-06-17`, `gemini-2.0-flash-lite-001`, `gemini-2.0-flash-001`, `gemini-1.5-flash-002`, `gemini-1.5-flash-001`, `gemini-1.5-pro-002`, `gemini-1.5-pro-001`, `claude-sonnet-4@20250514`, `claude-sonnet-4-5@20250929`, `claude-haiku-4-5@20251001`, `claude-3-7-sonnet@20250219`, `claude-3-5-sonnet@20240620`, `claude-3-5-sonnet-v2@20241022`, `claude-3-haiku@20240307`, `gpt-5-2025-08-07`, `gpt-5.1-2025-11-13`, `gpt-5.2-2025-12-11`, `gpt-5.4-2026-03-05`, `gpt-5.4-mini-2026-03-17`, `gpt-5.4-nano-2026-03-17`, `gpt-5.5-2026-04-23`, `gpt-5-mini-2025-08-07`, `gpt-5-nano-2025-08-07`, `gpt-4.1-2025-04-14`, `gpt-4.1-mini-2025-04-14`, `gpt-4.1-nano-2025-04-14`, `gpt-4o-mini-2024-07-18`, `gpt-4o-2024-11-20`, `gpt-4o-2024-08-06`, `gpt-4o-2024-05-13`, `gpt-4-0613`, `gpt-4-0314`, `gpt-4-turbo-2024-04-09`, `gpt-3.5-turbo-0125`, `gpt-3.5-turbo-1106`, `watt-tool-8b`, `watt-tool-70b`
- `tool_ids` (list of string, optional, nullable) — Eine Liste der IDs von Tools, die der Agent verwendet
- `tool_ids_strict` (boolean, optional, nullable) — Gilt nur als Überschreibung für eine Unterhaltung und ist zulässig, wenn tool_ids überschrieben werden dürfen. Wenn true, enthält die überschreibende tool_ids die vollständige Liste der in der Unterhaltung erlaubten Tools, einschließlich der Tools in Workflow-Knoten. Wird ignoriert, wenn es für den Agenten festgelegt ist.
- `native_mcp_server_ids` (list of string, optional, nullable) — Eine Liste der nativen MCP-Server-IDs, die der Agent verwenden soll
- `knowledge_base` (list of KnowledgeBaseLocator, optional, nullable) — Eine Liste der Wissensdatenbanken, die der Agent verwenden soll

### KnowledgeBaseRagChunkModel

- `chunk_id` (string, required)
- `document_id` (string, required)
- `content` (string, required)

### TransferToAgentToolResultSuccessModelOutputBranchInfo

- `branch_reason`: `configured` (TransferBranchInfoConfigured)
  - `branch_id` (string, required)
- `branch_reason`: `defaulting_to_main` (TransferBranchInfoDefaultingToMain)
  - `branch_id` (string, required)
- `branch_reason`: `traffic_split` (TransferBranchInfoTrafficSplit)
  - `branch_id` (string, required)
  - `traffic_percentage` (double, required)

### WorkflowToolResponseModelOutputStepsItems

- `type`: `edge` (WorkflowToolEdgeStepModel)
  - `edge_id` (string, required)
  - `step_latency_secs` (double, required)
  - `target_node_id` (string, required)
- `type`: `max_iterations_exceeded` (WorkflowToolMaxIterationsExceededStepModel)
  - `max_iterations` (integer, required)
  - `step_latency_secs` (double, required)
- `type`: `nested_tools` (WorkflowToolNestedToolsStepModel)
  - `is_successful` (boolean, required)
  - `node_id` (string, required)
  - `requests` (list of ConversationHistoryTranscriptToolCallCommonModel-Output, required)
  - `results` (list of WorkflowToolResponseModelOutputStepsItemsDiscriminatorMappingNestedToolsResultsItems, required)
  - `step_latency_secs` (double, required)

### KnowledgeBaseLocator

- `type` (enum, required) — Der Typ der Wissensdatenbank
  - Allowed values: `file`, `url`, `text`, `folder`
- `name` (string, required) — Der Name der Wissensdatenbank
- `id` (string, required) — Die ID der Wissensdatenbank
- `usage_mode` (enum, optional, default: auto) — Der Nutzungsmodus der Wissensdatenbank
  - Allowed values: `prompt`, `auto`

### WorkflowToolResponseModelOutputStepsItemsDiscriminatorMappingNestedToolsResultsItems

## Examples

**Response**

```json
{
  "agent_id": "agent_3701k3ttaq12ewp8b7qv5rfyszkz",
  "status": "processing",
  "metadata": {
    "start_time_unix_secs": 1714423232,
    "call_duration_secs": 10,
    "cost_fiat": 1.1
  },
  "conversation_id": "conv_7401k5m9x2p8ec3rqv6dtnhb0fzw",
  "has_audio": true,
  "has_user_audio": true,
  "has_response_audio": true,
  "has_auxiliary_audio": true,
  "transcript": [
    {
      "role": "user",
      "time_in_call_secs": 10,
      "message": "Hello, how are you?"
    }
  ],
  "agent_name": "My agent",
  "version_id": "agtvrsn_5xM3yVvZQKV0EfqQpLr2",
  "environment": "production",
  "tag_ids": []
}
```

**SDK Code**

```python
import requests

url = "https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM"

response = requests.get(url)

print(response.json())
```

```javascript
const url = 'https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM';
const options = {method: 'GET'};

try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}
```

```go
package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM"

	req, _ := http.NewRequest("GET", url, nil)

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(res)
	fmt.Println(string(body))

}
```

```ruby
require 'uri'
require 'net/http'

url = URI("https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)

response = http.request(request)
puts response.read_body
```

```java
import com.mashape.unirest.http.HttpResponse;
import com.mashape.unirest.http.Unirest;

HttpResponse<String> response = Unirest.get("https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM")
  .asString();
```

```php
<?php
require_once('vendor/autoload.php');

$client = new \GuzzleHttp\Client();

$response = $client->request('GET', 'https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM');

echo $response->getBody();
```

```csharp
using RestSharp;

var client = new RestClient("https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM");
var request = new RestRequest(Method.GET);
IRestResponse response = client.Execute(request);
```

```swift
import Foundation

let request = NSMutableURLRequest(url: NSURL(string: "https://api.elevenlabs.io/v1/convai/conversations/21m00Tcm4TlvDq8ikWAM")! as URL,
                                        cachePolicy: .useProtocolCachePolicy,
                                    timeoutInterval: 10.0)
request.httpMethod = "GET"

let session = URLSession.shared
let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in
  if (error != nil) {
    print(error as Any)
  } else {
    let httpResponse = response as? HTTPURLResponse
    print(httpResponse)
  }
})

dataTask.resume()
```