ElevenLabs Alternatives: Open-Source and Hosted Agent Platforms

For voice agents, the ElevenLabs alternatives that matter replace ElevenAgents, its hosted agent platform. Vapi, Retell AI, Bland AI, Synthflow and Smallest AI are hosted and closed source. Dograh, the open-source AI voice agent platform we build, can keep ElevenLabs voices with your own key, and you can self-host it or run it on Dograh Cloud.

Key Takeaways

• ElevenLabs voices and Scribe speech-to-text run inside Dograh with your own key.
• ElevenAgents private deployment runs in your own AWS or GCP account, through sales.
• The $0.08 agent minute leaves out the language model and the phone line.

This guide sorts out what ElevenLabs sells, then sets the agent platforms that replace ElevenAgents against it on license, hosting, models and list price in one table, with Dograh right after it. Later sections cover voice-only swaps, what ElevenAgents costs, when staying wins and what moving off takes.

What ElevenLabs sells, and why teams look elsewhere

ElevenLabs sells three products, and only one of them is an agent platform. ElevenCreative is the credit plan creators use for voiceovers. ElevenAPI sells the speech models by the character or by the hour, from Flash to Eleven v4 for voices and Scribe for speech-to-text. ElevenAgents is the hosted platform that turns those models into phone and web agents.

It is a large, serious vendor. TechCrunch reported in May 2026 that ElevenLabs had passed $500 million in annual recurring revenue, after a $500 million Series D at an $11 billion valuation. Its voices also lead on quality. Eleven v4 Turbo and Eleven v4 hold the top two places on the Artificial Analysis Speech Arena, where listeners vote blind between voices.

ElevenAgents wraps those voices in a full product, with ElevenLabs' own speech-to-text and turn-taking, a choice of language models, Realtime speech-to-speech models, visual Workflows, tests, native Twilio and SIP trunking.

Teams start looking at ElevenLabs competitors for concrete reasons. The platform code is closed, and its public GitHub holds SDKs and starter kits. Inside an agent, the voice and the transcription come from ElevenLabs' own models. Your own language model plugs in through an OpenAI-compatible endpoint that ElevenLabs' cloud calls, and it cannot be paired with the Realtime models.

Hosting is the other trigger. Agents run on ElevenLabs' cloud, with data in the U.S. by default. EU, India or Singapore data residency is an Enterprise feature. So is a business associate agreement (BAA) under the Health Insurance Portability and Accountability Act (HIPAA), and patient data also needs Zero Retention Mode turned on. ElevenLabs' private deployment runs ElevenAgents inside your own AWS or GCP account, arranged through its sales team under an NDA, with its v2 and v2.5 voices and Scribe V2. The newest Eleven v3 and v4 voices are not part of it.

How we compared the alternatives

Four questions sort the field. The license decides whether you can read the code and keep running it if the vendor changes course. Where it runs means your own servers without a sales contract, or the vendor's cloud only. Model choice covers bringing your own keys and hosting open-weight models. Last, a full platform ships the builder, telephony, call history and QA, while a framework leaves the product to your developers.

List prices come from each vendor's own pricing page. Latency is left out, because it depends on your models and phone path far more than on the platform. Voiceover and text-to-speech tools such as Murf and Cartesia change only the voice, so they get their own short section below.

ElevenLabs alternatives in one table

Among the open-source options, Dograh is the one whose open code includes the visual builder, telephony, call traces and QA. Bolna's open core has no dashboard, and LiveKit Agents and Pipecat are frameworks.

Dograh: keep the voices, own the platform

Dograh's code is open under BSD-2-Clause, and its GitHub repository holds the whole platform, including Campaigns for outbound lists, HTTP API tools that read and write your systems during the call, call traces and automatic QA.

If your team likes ElevenLabs' voices, you do not have to give them up. ElevenLabs is a built-in voice provider in Dograh, from Flash v2.5, the default, up to Eleven v4 and v4 Turbo. You paste the voice ID from your Voice Library, cloned voices included, and add your own ElevenLabs key. Scribe v2 Realtime is there as the speech-to-text option too, and the voice settings let you point the base URL at an ElevenLabs Data Residency endpoint. We built a page on keeping your ElevenLabs voices on a platform you own, from connecting your key to hearing the voice on a real phone line.

Dograh is the orchestration layer that runs the call flow, tools and telephony, and the speech and language models are a separate layer you choose. Next to ElevenLabs sit voice providers such as Cartesia, Deepgram and Rime, Dograh's own models, and local voices such as Kokoro through Speaches, an OpenAI-compatible speech server. When you bring your own key (BYOK), ElevenLabs still processes the speech it voices or transcribes, as our post on BYOK vs managed models explains. Your ElevenLabs voice also fits Dograh's pre-recorded clip pattern, where the language model plays a recorded line when one fits and falls back to a cloned version of the same voice for everything else.

Agents are built in the visual workflow builder or from an AI assistant through Dograh's MCP server. Dograh connects to your own carrier, such as Twilio, Vonage, Plivo or Telnyx, or to Dograh SIP. When a call needs a person, Transfer Call works on Twilio, Telnyx, Plivo, Cloudonix and Asterisk ARI. Dograh calls the destination first and connects the caller only when someone answers. On Twilio, it can play a short spoken introduction to both people before connecting them.

What it leaves to you is just as concrete. There is no keypad (DTMF) tool and no built-in SMS, and CRMs connect through HTTP API tools and webhooks rather than native integrations. Self-hosting with Docker Compose runs 10 calls at once by default, and the servers and their upkeep sit with your team. Self-hosted Dograh has no per-minute platform fee (servers, model keys and phone minutes still cost); Dograh Cloud has a 1¢-a-minute platform fee. Our Dograh vs ElevenLabs comparison goes through the two row by row.

Open Source Alternative to Vapi / Retell

Self-hosted voice agent platform — no per-minute fees

dograh-hq/dograh

Star on GitHub

Hosted alternatives to ElevenAgents

Five hosted platforms come up most often, and like ElevenAgents, each runs on the vendor's cloud with closed code.

Vapi and Retell AI

Vapi is a developer platform with a $0.05-a-minute hosting fee, with models and phone billed on top, and HIPAA is a $2,000-a-month add-on. Vapi retired its visual Workflows on 18 August 2026, so building there means Assistants and Squads, closer to code than ElevenLabs' Workflows.

Retell AI publishes $0.07 to $0.31 a minute for a full agent, from $0.055 for its voice infrastructure. Every account gets 20 concurrent calls, and each extra one is $8 a month. It has a visual Conversation Flow builder and sells OpenAI's realtime speech-to-speech models.

Bland AI and Synthflow

Bland AI runs on its own speech and language models, and one per-minute rate covers them: $0.14 a minute on Start, with 100 calls a day, or $0.12 plus $299 a month on Build. On-prem or VPC deployment and a BAA come on Enterprise. Synthflow is sold as an annual enterprise agreement, priced by call volume and setup, with no free plan or per-minute rate.

Smallest AI and Bolna

Smallest AI is a model company with its own hosted agent platform, Atoms, at $0.09 to $0.21 a minute on Pay as you go, depending on the models. Like ElevenLabs, it pairs its own models with a hosted platform, and our guide to Smallest AI alternatives makes the same keep-the-voices point.

Bolna's orchestration core is open source under MIT. Its standard rate is 6¢ a minute with its preferred models included, plus telephony and a platform fee per minute, and it is built for India, with more than 10 Indian languages and 20+ languages in all. Our guide to Bolna alternatives covers its field.

Open-source routes: LiveKit Agents and Pipecat

ElevenLabs' private deployment is licensed software running in your cloud account, while open source lets anyone read the code and run it without a contract. LiveKit Agents (Apache-2.0) and Pipecat (BSD-2-Clause) are the two open-source frameworks teams weigh most, and both can call ElevenLabs voices. Pipecat is a Python framework from Daily, and Dograh itself is built on a fork of it. In our experience it is an excellent base, but production adds phone integrations, call history, QA on live calls and the call lifecycle, which is the part Dograh ships.

Our comparison of the best open-source voice AI platforms goes deeper on the frameworks. If open source is your main reason for leaving, our post on the open-source alternative to ElevenLabs sorts open voices, open frameworks and an open platform.

Swapping only the voice

Some teams only want to change the voice. Speech APIs such as Cartesia and Rime replace ElevenLabs' text to speech. Deepgram sells speech-to-text and voices, and also a Voice Agent API that bundles speech-to-text, a language model and a voice in one connection. Our guide to Deepgram alternatives covers both of its layers.

Open-weight voices are the other route. Kokoro is an 82-million-parameter model under Apache-2.0 with fixed voices and no cloning. Chatterbox is MIT-licensed and clones a voice from a short clip. Both sit well below ElevenLabs on the Speech Arena, outside its top 60, so test them on your own calls before you commit. None of these replaces ElevenAgents on its own, though all three speech APIs are built-in providers in Dograh and Kokoro runs there through Speaches. Chatterbox is a model to test behind an OpenAI-compatible server.

What ElevenAgents costs

ElevenAgents bills by the call minute, and every paid plan works out to $0.08 a minute. ElevenLabs' agent pricing runs from a free plan with 15 minutes and 4 concurrent calls to Business at $990 a month for 12,375 minutes and 40 concurrent calls. Extra minutes cost $0.08, and burst pricing charges $0.16 a minute for calls above your concurrency limit.

The $0.08 covers ElevenLabs' own speech models in one agent pipeline. The language model is billed on top at the provider's published rate, and ElevenLabs charges nothing for telephony, so your Twilio account or SIP carrier bills you directly. Our breakdown of ElevenLabs pricing works through the plans and a sample month.

When staying on ElevenLabs makes sense

If you want the top-ranked voices and the agent from one vendor on one bill, with no servers to run, ElevenAgents gives you that. It also covers things Dograh does not. Callers can answer with the keypad, and the agent can press keys itself. Transfers can be blind, conference or SIP REFER, and burst capacity lets a plan handle up to three times its usual concurrency. Built-in tests, A/B experiments and SDKs for mobile and Unity round it out.

Buyers who need a vendor BAA and EU, India or Singapore residency get them on Enterprise. A self-hosted Dograh install sits inside your own perimeter instead, so no BAA with Dograh is needed, though ElevenLabs needs one, on Enterprise, if its voices handle patient data, and the phone carrier may too, depending on what it stores.

What moving off ElevenAgents takes

Moving can happen one number at a time, with ElevenAgents answering the rest until each rebuilt agent passes its tests. There is no ElevenLabs importer in Dograh, so treat each agent's prompt and Workflow as the spec. Rebuild it by hand in the builder, or have an AI assistant draft it through Dograh's MCP server, then check every node yourself.

The voice can come with you, as the same voice ID under your own key. Numbers already sit with your carrier, because ElevenAgents uses your own Twilio account or SIP trunk, so point each one at Dograh when its agent is ready. Flows that collect keypad digits need a voice answer instead, and every rebuilt agent should be tested on a real phone line, because browser test calls cannot transfer.

Picking your ElevenLabs alternative

Teams that would rather write the agent in code can start with LiveKit Agents or Pipecat. On a hosted plan, Vapi suits developers who want an API and their own keys, and Retell AI suits teams that want a drawn flow with a rate card. Bland AI fits teams happy for the vendor to pick every model, and Synthflow a vendor-led launch under one contract. If you want the platform's code and your own choice of models, Dograh answers the reasons that start most of these searches. Run it from GitHub with Docker Compose, or start on Dograh Cloud. Whichever you shortlist, run the same call script on a real phone line, since most phone calls still reach the agent as 8 kHz audio while a browser demo runs at 16 kHz.

Each part of this guide has its own deep dive:

• Dograh vs ElevenLabs
• Open-Source Alternative to ElevenLabs
• ElevenLabs Pricing Explained