GET STARTED

Changelog

November 5, 2025

UI

  • Improved notifications: Updated notification system to display notifications relevant to the active platform you’re viewing, providing a more focused and contextual experience.

Agents Platform

  • Dynamic variable transfer destinations: Agent transfers now support dynamic variables for phone numbers and SIP URIs, enabling runtime-determined transfer destinations based on conversation context.
  • MCP tool configuration overrides: Added the ability to create, update, retrieve, and delete custom configuration overrides for specific MCP Server tools, allowing fine-grained control over tool behavior and parameters.

Text to Dialogue

  • Timestamps and voice segments: Text to Dialogue now supports timestamped outputs with character-level alignment and voice segment tracking, making it easier to synchronize dialogue with animations or subtitles.

Music API

  • Stem separation: Added new stem separation endpoint to isolate different audio components (vocals, drums, bass, instruments) from existing music tracks.
  • Increased prompt length: Music generation now supports prompts up to 4,100 characters, with individual lyric lines supporting up to 200 characters.

Security

  • Single-use tokens: Introduced time-limited single-use token generation for secure operations, providing enhanced security for sensitive API operations.

SDK Releases

JavaScript SDK

  • v2.22.0 - Updated with latest API schema changes including workspace model approvals and MCP tool configuration endpoints.
  • v2.21.0 - Added support for intercepting raw WebSocket messages via a general message handler, allowing developers to access all messages beyond the standard callbacks, including undocumented message types like agent_tool_response.

Python SDK

  • v2.22.0 - Updated with latest API schema changes including workspace model approvals and MCP tool configuration endpoints.
  • v2.21.0 - Replaced print statements with proper logging to support better debugging and production use cases.

Agents Packages

Android SDK

  • v0.4.0 - Improved type safety by using ConversationMode enums instead of strings, migrated from LiveData to StateFlow with backward compatibility support, and fixed bug where agents wouldn’t end calls when requested.

iOS SDK

  • v2.0.16 - Fixed issue where the end_call tool wasn’t properly ending conversations, ensuring agents can correctly terminate calls.

CLI

  • @elevenlabs/cli@0.2.0 - Removed --env flag support for virtual environment isolation. This feature will be reintroduced once proper environment isolation is supported in the product.

API

TLS

Our TLS endpoints no longer allow some older and insecure cipher modes. Most clients should not be affected as they already negotiate a modern cipher mode.

New Endpoints

Agents Platform

MCP Tool Configuration

Text to Dialogue

Tokens

Music API

  • POST /v1/music/stem-separation - Separate audio into individual stems (vocals, drums, bass, instruments). Accepts multipart file upload and returns a ZIP archive.

Updated Endpoints

Text to Dialogue

  • Text to dialogue with timestamps

    • Added voice_segments array with required dialogue_input_index field to track which dialogue input each segment corresponds to
    • Enhanced response schema with character-level alignment data

  • Stream text to dialogue with timestamps

    • Added streaming support for voice_segments with character-level alignment
    • Includes dialogue_input_index in voice segment chunks

Music

  • Music prompt endpoints updated:
    • Maximum prompt length increased from previous limit to 4,100 characters
    • Lyrics lines now support up to 200 characters per line

Workspace

  • Get workspace resource
    • Added dashboard and dashboard_configuration to WorkspaceResourceType enum
    • Response schema changes (breaking compatibility - see migration notes below)

Billing

  • Get subscription
    • Added next_billing_period field (required) to PendingSubscriptionSwitchResponseModel
    • Added subtotal_cents and tax_cents (nullable) to InvoiceResponseModel for better invoice breakdown

Agents Platform

Batch Calling

  • Submit batch calling

    • Made agent_phone_number_id nullable and optional
    • Made phone_number nullable and optional in recipient models
    • Response schema changes (breaking compatibility - see migration notes below)

  • Get batch calling status

    • Made phone_number_id and phone_provider nullable

  • Cancel batch calling & Retry batch calling

    • Made phone-related fields nullable
    • Response schema changes (breaking compatibility - see migration notes below)

Transfer Destinations

  • Transfer-related endpoints now support dynamic variable transfer destinations:
    • Added PhoneNumberDynamicVariableTransferDestination type for phone number transfers using dynamic variables
    • Added SIPUriDynamicVariableTransferDestination type for SIP URI transfers using dynamic variables
    • Updated PhoneNumberTransfer discriminators to include dynamic variable types

MCP Servers

  • Create MCP server, Get MCP server, Update MCP server
    • Added tool_config_overrides field for per-tool configuration customization
    • Added request_headers support in MCPServerConfigUpdateRequestModel
    • Added tool_call_sound and tool_call_sound_behavior configuration options

Tools

  • Tool schema enhancements:
    • LiteralJsonSchemaProperty now includes is_system_provided flag
    • Agent tool headers now support ConvAIDynamicVariable for dynamic header values

Twilio Integration

  • Phone number endpoints updated:
    • Added optional region_config field for Twilio phone numbers
    • Added TwilioRegionId enum for region selection
    • Added TwilioEdgeLocation enum for edge location configuration

Agents