ElevenLabs Agents

Summary

ElevenLabs Agents is ElevenLabs' hosted platform for building conversational voice and chat agents in 70+ languages. It coordinates a fine-tuned speech-to-text model, your choice of LLM, a low-latency text-to-speech model and a proprietary turn-taking model in one pipeline.

Best for: Developers and product teams who want ElevenLabs' voice quality and cloning as the foundation of their agent, and are comfortable assembling the rest of the stack themselves.
Not for: Non-technical ops teams wanting a turnkey, dashboard-configured agent with built-in CRM/helpdesk integrations and per-resolution pricing.

Captured 2026-09-08 · theagentsindex.com

From: Freethen $6 monthly · $5 billed annually
Free tier: YesFree plan at $0 per month: 15 minutes of calls included, 4 concurrent calls, workflow builder, knowledge base, multilingual and widget. 1 Full Seat and 20 Basic Seats; no commercial licence (that starts on Starter) and no additional-minute purchase.

Visit ElevenLabs Agents Compare Find alternatives

What it does

Writes code: PartlyElevenLabs agents do not author code themselves; the Cursor integration lets an agent launch Cursor background agents that do, and needs a paid Cursor plan.Start a new Cursor background agent to work on a GitHub repository. The agent can add features, fix bugs, refactor code, and optionally create a pull…elevenlabs.io/docs/eleven-agents/custom…
Debugs and fixes: PartlyBug fixing is delegated: the agent starts and steers a Cursor background agent, which diagnoses and lands the change; ElevenLabs itself never touches the repository.Start a new Cursor background agent to work on a GitHub repository. The agent can add features, fix bugs, refactor code, and optionally create a pull…elevenlabs.io/docs/eleven-agents/custom…
Ships it: PartlyThe CLI pushes agent configuration to the platform and into CI/CD; what it deploys is agents, not application code.Set up CI/CD integration to automatically deploy your agentselevenlabs.io/docs/eleven-agents/operat…
Done-for-you service: PartlySelf-serve software you run, but enterprises can buy ElevenLabs Forward Deployed Engineers who scope, build and stabilise the deployment; no price is published.Our Forward Deployed Engineers work alongside your team to scope, architect, and launch production-ready AI systems.elevenlabs.io/agents/forward-deployed-e…

How much it does unattended

Runs autonomously: YesAgents run whole calls and chats unattended, including scheduled outbound campaigns, with explicit system tools to end a call or hand off to a human number.Batch Calling enables you to initiate multiple outbound calls simultaneously using your configured ElevenLabs agents.elevenlabs.io/docs/eleven-agents/phone-…
Multi-agent: YesAn orchestrator agent routes a live conversation to specialist agents on natural-language conditions, and transfers can be nested; the transcript carries across.Agent-to-agent transfer allows an ElevenLabs agent to hand off the ongoing conversation to another designated agent when specific conditions are met.elevenlabs.io/docs/eleven-agents/custom…
Agent permissions: YesTool use is gated per MCP server and per tool: always ask, fine-grained (auto-approved, requires approval, disabled), or no approval at all.Always Ask (Recommended): Maximum security. The agent will request your permission before each tool use.elevenlabs.io/docs/eleven-agents/custom…

Which models it runs on

Claude: YesAnthropic models are natively selectable in agent settings; a few older Claude models drop out in EU data-residency workspaces.for example, OpenAI's GPT-5.6 / GPT-5.5 series, Anthropic's Claude Opus and Claude Sonnet 5, or Google's Gemini Pro modelselevenlabs.io/docs/eleven-agents/custom…
GPT: YesOpenAI models are natively supported and are the one provider family the vendor says stays fully available under EU data residency.Custom LLMs and OpenAI LLMs remain fully available.elevenlabs.io/docs/eleven-agents/custom…
Open models: YesOpen-weight models run two ways: ElevenLabs-hosted Qwen in the native model list, or Llama and Mixtral through a custom LLM endpoint such as Together AI, Groq or SambaNova.it's recommended to use at least meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo for your ElevenLabs Agents agent.elevenlabs.io/docs/eleven-agents/custom…
Model choice: YesYou pick the model per agent, plus a backup cascade; LLM usage is billed on top of the plan and varies by model.ElevenAgents provides a unified interface to connect your agent to multiple models and providers, offering flexibility, reliability, and cost…elevenlabs.io/docs/eleven-agents/custom…

Where you use it

In your editor: NoThe vendor enumerates its surfaces as web widgets, mobile apps, phone systems and custom integrations; no editor or IDE extension appears in that list.The Integrate section provides everything you need to connect your agents to your users, whether through web widgets, mobile apps, phone systems, or…elevenlabs.io/docs/eleven-agents/integr…
On the command line: YesA standalone CLI authenticates by OAuth and pushes, pulls and diffs agent, tool and test configurations held as files in version control.The ElevenLabs CLI allows you to access your ElevenLabs agents from your terminal, unlocking new ways to manage themelevenlabs.io/docs/eleven-agents/operat…
In your pipeline: YesAgents deploy from CI with an API key in the environment; the unit of change is the agent configuration, not a pull request in your code.Set up CI/CD integration to automatically deploy your agentselevenlabs.io/docs/eleven-agents/operat…
In a browser: YesAgents are built, monitored and A/B tested in a hosted dashboard, and shipped to end users as an embeddable web widget.Build your first conversational AI agent in just a few minutes using our web platform. No coding required.elevenlabs.io/conversational-ai

Whose machine it runs on

Self-hosted: PartlyThe full agent runtime can run in your own VPC or hardware on AWS or GCP, but the documentation is NDA-gated and sold through the account team; Azure is unreleased.ElevenAgents runs full conversational agent workflows inside your own cloud account or your own hardware.elevenlabs.io/docs/overview/capabilitie…
Open source: PartlyThe CLI and the client SDKs are public on GitHub and take contributions; the agent platform and the speech models themselves are proprietary and hosted.Find the complete source code and contribute on GitHub.elevenlabs.io/docs/eleven-agents/operat…

What it costs to run

How it meters: YesA monthly plan buys included call minutes and concurrency; overage is per minute, texts are per message, and LLM plus telephony are billed on top at usage.Additional call minutes cost $0.08 per minute (burst pricing $0.16 per minute, double the standard rate, when you exceed your concurrency limit), and…elevenlabs.io/pricing/agents
Free tier: YesA permanent $0 plan includes 15 call minutes a month, four concurrent calls, the workflow builder, knowledge base and widget; no sales call.Free $0 per month Build for free 15 minutes of calls included 4 Concurrent Callselevenlabs.io/pricing/agents
API access: YesAgents are created, driven and monitored over REST and a WebSocket protocol, with first-party SDKs for web, Python, iOS, Android and React Native.Yes, we have a WebSocket API and Javascript, React, Python and iOS SDKs.elevenlabs.io/conversational-ai
MCP server: YesA hosted, OAuth-authenticated MCP server lets another assistant create, edit, cost-estimate and read the transcripts of your agents; per-region URLs exist for isolated workspaces.The ElevenLabs hosted MCP server is a remote Model Context Protocol (MCP) server that exposes agent management tools to AI assistants.elevenlabs.io/docs/eleven-agents/operat…
Bring your own key: YesPoint an agent at your own OpenAI-compatible endpoint with your own key held in workspace secret storage; inference then bills to you, not to ElevenLabs credits.Custom LLMs let you bring your own OpenAI API key or run an entirely custom LLM server.elevenlabs.io/docs/eleven-agents/custom…

Buying it for a team

A company can buy it: YesWorkspaces are the unit of purchase, with self-serve card payment up to Business and a contract, DPA and SLA path on Enterprise.ElevenLabs workspaces use a seat-based model. Every member in a workspace occupies one seat, and the seat type they hold determines what they can…elevenlabs.io/docs/overview/administrat…
Seat model: YesTwo seat types with a hard ceiling: Full Seats are capped at 11 per self-serve workspace, every plan includes 20 Basic Seats, and Enterprise seat counts sit in the contract.Pro, Scale, and Business workspaces can purchase additional Full Seats up to a maximum of 11 Full Seats per workspace, including seats included with…elevenlabs.io/docs/overview/administrat…
Pooled budget: YesOne shared workspace balance for everyone; admins carve it up with billing groups that set credit quotas per team and reject requests once a group is exhausted.Every user in a workspace draws from the same shared credit pool — Basic Seat users do not receive their own credit allocation.elevenlabs.io/docs/overview/administrat…
Admin controls: PartlyGuardrails constrain what an agent may say and per-tool approval constrains what it may do, but Guardrails is in Alpha and enabling MCP servers is not restricted to admins.Guardrails give teams a powerful way to govern how agents behave in production, keeping them on-topic, on-brand, and resistant to manipulation at…elevenlabs.io/docs/eleven-agents/best-p…
Audit log: PartlyOCSF-schema logs over 100 administrative endpoints, pullable into a SIEM, but gated on Enterprise tier plus an API key holding audit_log_read.Audit logs provide a chronological record of administrative actions performed within your workspace, supporting security monitoring, compliance…elevenlabs.io/docs/overview/administrat…
Single sign-on: PartlySAML and OIDC with domain verification, plus SCIM provisioning, all Enterprise-only; SAML is service-provider-initiated only and Entra over OIDC is discouraged.SSO is currently only available for Enterprise customers, and only Workspace admins can enable this feature.elevenlabs.io/docs/overview/administrat…

What happens to your code

Opt out of training: YesAny account can switch off model training in Terms and privacy, Data use; Enterprise data is excluded by default, and third-party LLM providers are contractually barred from training on it.Anyone can opt out of data use at any time by following the instructions below.elevenlabs.io/docs/help-center/legal/is…
Data residency: PartlyUS by default; EU, India and Singapore exist as separate Enterprise workspaces with their own accounts, API hosts and reduced model choice, and processing may still leave the region.As a standard, ElevenLabs' customer data is hosted/stored in the U.S., however ElevenLabs has released additional storage locations in the EU, India,…elevenlabs.io/docs/overview/administrat…
Getting out: PartlyAgents, tools and tests pull down as files via the CLI and conversations are readable over the API, but unused paid credits expire on cancel or downgrade and rollover caps at two months.After that, your account will be downgraded to the Free plan and any unused paid credits will expire.elevenlabs.io/pricing/agents
Certifications: YesSOC 2, ISO 27001, GDPR and HIPAA are all claimed, with BAAs for HIPAA customers listed on the Enterprise plan and zero-retention available as a configuration.We adhere to SOC 2, ISO 27001, GDPR, and HIPAA standards. Deployments can be configured with zero-data-retention to meet your security, privacy, and…elevenlabs.io/agents/forward-deployed-e…

32 sourced claims on this page. Checked 2026-09-02. How we check.

Alternatives

Where we would send a reader instead, in that listing's own words.

Retell AI
Platform for building, testing and deploying AI voice and chat agents on phone, web and SMS, billed per minute.
Bland
Enterprise voice AI platform for phone agents, running its own self-hosted models.
OpenAI Realtime API
OpenAI's speech-to-speech API for low-latency voice agents over WebRTC, WebSocket or SIP.
Cartesia
Real-time speech models and a managed runtime for production phone and web voice agents.

Filed under

APIAutonomousClaudeEnterpriseFree tierGPTMulti-agentNo-codeOpen models

Sources and updates

Pages we read on elevenlabs.io23

Every fact below comes from the vendor. Nothing here is independently corroborated yet.

Coverage elsewhere2

Updates3

2026-09-02Capabilities · Commercial terms · Free tier
2026-08-31Capabilities · Commercial terms · Free tier · Price
2026-08-22Summary