> ## Documentation Index > Fetch the complete documentation index at: https://docs.oneinbox.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create Call ## OpenAPI ````yaml /openapi/openapi.yaml post /v1/calls openapi: 3.1.0 info: title: OneInbox API version: 1.0.0 servers: - url: https://api-tokyo.oneinbox.ai description: Production (Tokyo ap-northeast-1) security: [] tags: - name: health - name: integrations - name: llm-models - name: tools - name: knowledge-bases - name: voices - name: webhooks - name: agents - name: phone-numbers - name: calls - name: workspace - name: usage - name: audit-events paths: /v1/calls: post: tags: - calls summary: Create Call operationId: create_call_v1_calls_post requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/OutboundCallCreate' responses: '201': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/CallResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - API Key: [] components: schemas: OutboundCallCreate: properties: agent_id: type: string format: uuid title: Agent Id to_number: type: string pattern: ^\+[1-9]\d{1,14}$ title: To Number from_number: type: string pattern: ^\+[1-9]\d{1,14}$ title: From Number scheduled_at: anyOf: - type: string format: date-time - type: 'null' title: Scheduled At variables: additionalProperties: true type: object title: Variables webhook_url: anyOf: - type: string - type: 'null' title: Webhook Url overrides: anyOf: - $ref: '#/components/schemas/CallOverrides' - type: 'null' metadata: additionalProperties: true type: object title: Metadata use_shisa: type: boolean title: Use Shisa default: false type: object required: - agent_id - to_number - from_number title: OutboundCallCreate CallResponse: properties: id: type: string title: Id agent_id: type: string title: Agent Id type: type: string title: Type status: type: string title: Status from_number: anyOf: - type: string - type: 'null' title: From Number to_number: anyOf: - type: string - type: 'null' title: To Number room_name: anyOf: - type: string - type: 'null' title: Room Name direction: type: string title: Direction twilio_call_sid: anyOf: - type: string - type: 'null' title: Twilio Call Sid scheduled_at: anyOf: - type: string format: date-time - type: 'null' title: Scheduled At variables: additionalProperties: true type: object title: Variables started_at: anyOf: - type: string format: date-time - type: 'null' title: Started At ended_at: anyOf: - type: string format: date-time - type: 'null' title: Ended At duration_seconds: anyOf: - type: number - type: 'null' title: Duration Seconds end_reason: anyOf: - type: string - type: 'null' title: End Reason outcome: anyOf: - type: string - type: 'null' title: Outcome transcript: anyOf: - type: string - type: 'null' title: Transcript messages: items: {} type: array title: Messages recording_url: anyOf: - type: string - type: 'null' title: Recording Url ai_summary: anyOf: - type: string - type: 'null' title: Ai Summary analysis: anyOf: - additionalProperties: true type: object - type: 'null' title: Analysis latency: anyOf: - additionalProperties: true type: object - type: 'null' title: Latency description: >- Per-call latency/observability (JSONB), null until the call completes a turn. Per-component aggregates `stt`/`llm_ttft`/`tts_ttfb`/`eou_delay` = {avg_ms, p50_ms, min_ms, max_ms, samples, provider} (provider stamped per segment so the log stays accurate if the agent is reconfigured). `events` = turn-tagged stream [{turn, type, ms}] where type is one of stt|llm_ttft|tts_ttfb|eou_delay|kb|tool: — chains the whole response cycle per turn. `connect` = {setup_ms, dispatch_to_answer_ms, answer_to_active_ms, active_to_greeting_ms}. cost_cents: anyOf: - type: integer - type: 'null' title: Cost Cents credits_used: anyOf: - type: number - type: 'null' title: Credits Used kb_snapshot: anyOf: - additionalProperties: true type: object - type: 'null' title: Kb Snapshot metadata: additionalProperties: true type: object title: Metadata description: >- Caller-supplied metadata plus the platform key `concurrency_at_dispatch` (count of the org's calls in flight at dispatch, for load-test correlation). created_at: type: string format: date-time title: Created At updated_at: type: string format: date-time title: Updated At type: object required: - id - agent_id - type - status - from_number - to_number - room_name - direction - scheduled_at - variables - started_at - ended_at - duration_seconds - end_reason - transcript - messages - recording_url - analysis - latency - cost_cents - metadata - created_at - updated_at title: CallResponse HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError CallOverrides: properties: llm: anyOf: - $ref: '#/components/schemas/LLMOverride' - type: 'null' transcriber: anyOf: - $ref: '#/components/schemas/TranscriberOverride' - type: 'null' tts: anyOf: - $ref: '#/components/schemas/TTSOverride' - type: 'null' first_message: anyOf: - type: string - type: 'null' title: First Message type: object title: CallOverrides description: >- Per-call overrides — merged on top of the agent's stored config at dispatch. ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError LLMOverride: properties: provider: anyOf: - type: string enum: - openai - anthropic - groq - deepgram - shisa - custom - type: 'null' title: Provider model: anyOf: - type: string - type: 'null' title: Model temperature: anyOf: - type: number - type: 'null' title: Temperature system_prompt: anyOf: - type: string - type: 'null' title: System Prompt type: object title: LLMOverride TranscriberOverride: properties: provider: anyOf: - type: string enum: - deepgram - assembly_ai - whisper - azure - shisa - type: 'null' title: Provider model: anyOf: - type: string - type: 'null' title: Model language: anyOf: - type: string - type: 'null' title: Language type: object title: TranscriberOverride TTSOverride: properties: provider: anyOf: - type: string enum: - cartesia - deepgram - elevenlabs - openai - shisa - minimax - type: 'null' title: Provider voice_id: anyOf: - type: string - type: 'null' title: Voice Id model: anyOf: - type: string - type: 'null' title: Model speed: anyOf: - type: number - type: 'null' title: Speed stability: anyOf: - type: number - type: 'null' title: Stability type: object title: TTSOverride securitySchemes: API Key: type: http description: >- Your **API key** sent as a Bearer credential — `Authorization: Bearer `, where `` starts with `oi_sk_` (create one via `POST /v1/api-keys`). Although the field is labelled 'token', paste your API key here. A dashboard JWT is also accepted on this header for the same endpoints. scheme: bearer ````