For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Connect
BlogHelp CenterAPI PricingSign up
OverviewElevenCreativeElevenAgentsElevenAPIReception AIAPI referenceChangelog
OverviewElevenCreativeElevenAgentsElevenAPIReception AIAPI referenceChangelog
LogoLogo
Login
Login
Connect
BlogHelp CenterAPI PricingSign up

Changelog

October 7, 2025
October 7, 2025
Was this page helpful?
Previous

September 29, 2025

Next
Built with

Agents Platform

  • Gemini 2.5 Flash Preview models: Added support for gemini-2.5-flash-preview-09-2025 and gemini-2.5-flash-lite-preview-09-2025 LLM models, providing access to the latest September 2025 preview versions of Google’s Gemini 2.5 Flash models.
  • Claude Sonnet 4.5: Added support for claude-sonnet-4-5 and claude-sonnet-4-5@20250929 models, enabling access to the latest Claude Sonnet 4.5 model released on September 29, 2025.
  • Test invocations listing: Added new GET /v1/convai/test-invocations endpoint to list all test invocations with pagination support. Includes agent_id filter, page_size parameter (default 30, max 100), and cursor for pagination. Response includes test run counts, pass/fail statistics, and titles.
  • Agent archiving: Added archived field (boolean, default false) to agent platform settings, allowing agents to be archived without deletion while keeping them out of active agent lists.
  • MCP Server interruption control: Added disable_interruptions field (boolean, default false) to MCP server configuration, preventing user interruptions during tool execution for more reliable tool completion.
  • Streaming agent responses: Added agent_chat_response_part WebSocket event type for receiving partial agent chat responses in real-time during streaming conversations.
  • Workflow edge ordering: Added edge_order field (array of strings) to all workflow node types, enabling explicit control over edge evaluation order for deterministic workflow execution.
  • Test suite agent tracking: Added agent_id field (string, nullable) to test invocation responses for associating test runs with specific agents.

Voice Management

  • Voice generation source tracking: Added VoiceGeneration as a new source type in the History API for tracking audio generated from voice generation features.

Telephony

  • SIP trunk TLS validation: Added remote_domains field (array of strings, nullable) to SIP trunk configuration for specifying domains used in TLS certificate validation.

SDK Releases

JavaScript SDK

  • v2.18.0 - Updated with latest API schema changes from October 8, 2025

Python SDK

  • v2.17.0 - Updated with latest API schema changes and URL generation fixes from October 6, 2025

Packages

All packages updated with latest API schema changes:

  • @elevenlabs/react-native@0.3.2 - Updated TypeScript types and API client with new fields for agent archiving, MCP server configuration, and test invocations
  • @elevenlabs/react@0.7.1 - Updated React hooks and components with support for new agent settings and WebSocket events
  • @elevenlabs/client@0.7.1 - Core client library updated with new endpoint for test invocations listing and reorganized SDK method paths for secrets management
  • @elevenlabs/agents-cli@0.4.2 - CLI tool updated with support for new agent archiving flag and test invocation commands

MCP Server

  • v0.9.0 - Added option to return MCP server results as resource items for better integration with resource-based workflows

API

View API changes

New Endpoints

Agents Platform

  • GET /v1/convai/test-invocations - List all test invocations with pagination support
    • Parameters:
      • agent_id (required, string) - Filter by agent ID
      • page_size (optional, integer, default=30, max=100) - Number of results per page
      • cursor (optional, string) - Pagination cursor from previous response
    • Response: Returns paginated list with test run counts, pass/fail statistics, titles, and next cursor

New Fields

Agents Platform

  • Agent Settings: Added archived field (boolean, default false) to AgentPlatformSettingsRequestModel and AgentPlatformSettingsResponseModel for archiving agents
  • MCP Servers: Added disable_interruptions field (boolean, default false) to MCP server configuration schemas for preventing user interruptions during tool execution
  • Workflows: Added edge_order field (array of strings) to all workflow node types for explicit edge evaluation ordering
  • Test Invocations: Added agent_id field (string, nullable) to GetTestSuiteInvocationResponseModel for agent tracking

Telephony

  • SIP Trunks: Added remote_domains field (array of strings, nullable) to GetPhoneNumberInboundSIPTrunkConfigResponseModel and InboundSIPTrunkConfigRequestModel for TLS certificate validation

WebSocket Events

  • Added agent_chat_response_part to ServerEventType enum for streaming partial agent chat responses

Voice Management

  • Added VoiceGeneration to speech history source types

New LLM Models

Added the following models to the LLM enum:

  • claude-sonnet-4-5 - Claude Sonnet 4.5 latest
  • claude-sonnet-4-5@20250929 - Claude Sonnet 4.5 dated release (September 29, 2025)
  • gemini-2.5-flash-preview-09-2025 - Gemini 2.5 Flash preview (September 2025)
  • gemini-2.5-flash-lite-preview-09-2025 - Gemini 2.5 Flash Lite preview (September 2025)

Other Changes

Pronunciation Dictionaries

  • Updated parameter description for version_id in GET /v1/pronunciation-dictionaries/{dictionary_id}/{version_id}/download from “The id of the version of the pronunciation dictionary” to “The id of the pronunciation dictionary version”
  • Removed documentation note about UI limitation for multiple pronunciation dictionaries (multiple dictionaries now fully supported in UI)

Conversation History

  • Made type field optional in ConversationHistoryTranscriptOtherToolsResultCommonModel (previously required)