# Conversational voice-agent APIs

> 10 conversational voice agents ranked by the Anchor benchmark. Leader ElevenLabs Agents API + MCP (BB). Platforms that run a spoken conversation for you, either with one speech-to-speech model or a configurable pipeline of speech-to-text, a language model and text-to-speech. Compared on response latency, handling interruptions, tool calls, recovery after a misunderstanding and cost per completed call.

- Canonical: https://www.anchorterminal.com/categories/voice-agents
- Markdown: https://www.anchorterminal.com/categories/voice-agents.md (~2,800 tokens)
- Slim: https://www.anchorterminal.com/categories/voice-agents.min.md (~530 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/categories/voice-agents.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- Updated: 2026-10-04

Platforms that run a spoken conversation for you, either with one speech-to-speech model or a configurable pipeline of speech-to-text, a language model and text-to-speech. Compared on response latency, handling interruptions, tool calls, recovery after a misunderstanding and cost per completed call.

- Tools ranked: 10 · agent-ready (BB or better): 1 · accept x402: 0 · hosted endpoints: 10 · desk reviews by the panel: 20
- JSON: https://www.anchorterminal.com/api/v1/tools.json (list) · https://www.anchorterminal.com/api/v1/rankings.json (ranked) · https://www.anchorterminal.com/api/v1/x402.json (payable) · https://www.anchorterminal.com/api/v1/capabilities.json (by capability)
- Grades run AA, A, BB, B, C, D, E, F · methodology: https://www.anchorterminal.com/benchmark/

- Capabilities in this category: voice.agent, voice.speech-to-speech, voice.pipeline, voice.tools, voice.telephony
- https://letme.dev/voice.agent picks the top-graded tool in this list and says how to call it direct; calling through letme comes later (https://www.anchorterminal.com/letme/index.md)

## Ranking

| # | Tool | Vendor | Kind | Category | Grade | Score | Confidence | x402 | Auth | Where | Reviews | Page |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 83 | ElevenLabs Agents API + MCP | ElevenLabs | HTTP API | Voice agents | BB | 71.5 | medium | no | OAuth or key | hosted | 3.5/5 (2) | https://www.anchorterminal.com/tools/elevenlabs-agents.md |
| 114 | Retell AI API + MCP | Retell AI | HTTP API | Voice agents | B | 69.4 | medium | no | API key | hosted | 3.5/5 (2) | https://www.anchorterminal.com/tools/retell-ai.md |
| 126 | Deepgram Voice Agent API | Deepgram | HTTP API | Voice agents | B | 68.4 | medium | no | API key | hosted + local | 3/5 (2) | https://www.anchorterminal.com/tools/deepgram-voice-agent.md |
| 191 | Bland AI API + MCP | Bland AI | HTTP API | Voice agents | B | 64.1 | medium | no | API key | hosted + local | 3/5 (2) | https://www.anchorterminal.com/tools/bland-ai.md |
| 198 | Vapi API + MCP | Vapi | HTTP API | Voice agents | B | 63.7 | medium | no | API key | hosted + local | 2.5/5 (2) | https://www.anchorterminal.com/tools/vapi.md |
| 279 | Ultravox Realtime API | Ultravox (Fixie.ai) | HTTP API | Voice agents | C | 58.6 | medium | no | API key | hosted | 3/5 (2) | https://www.anchorterminal.com/tools/ultravox.md |
| 296 | Hume EVI (Empathic Voice Interface) | Hume AI | HTTP API | Voice agents | C | 57.3 | medium | no | OAuth or key | hosted | 2/5 (2) | https://www.anchorterminal.com/tools/hume-evi.md |
| 339 | Bolna API + MCP | Bolna | HTTP API | Voice agents | D | 52.9 | medium | no | API key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/bolna.md |
| 352 | Synthflow API + MCP | Synthflow | HTTP API | Voice agents | D | 51.3 | medium | no | OAuth or key | hosted | 2/5 (2) | https://www.anchorterminal.com/tools/synthflow.md |
| 383 | Vogent API | Vogent | HTTP API | Voice agents | D | 47.4 | medium | no | API key | hosted | 2/5 (2) | https://www.anchorterminal.com/tools/vogent.md |

Scores are from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/), with Performance and Task success pending. p95 latency and context cost come from our probes, which haven't run yet.

## Summaries

### 83. ElevenLabs Agents API + MCP, BB (71.5)

ElevenAgents (formerly Conversational AI) runs hosted voice agents as a pipeline of a fine-tuned ElevenLabs ASR model, an LLM of your choice or your own, ElevenLabs TTS and a proprietary turn-taking model. API keys scoped to endpoint groups, with per-key credit limits and service accounts. $0.08 a minute excludes LLM tokens and carrier minutes.

- Page: https://www.anchorterminal.com/tools/elevenlabs-agents · Markdown: https://www.anchorterminal.com/tools/elevenlabs-agents.md · JSON: https://www.anchorterminal.com/api/v1/tools/elevenlabs-agents.json
- Capabilities: voice.agent, voice.pipeline, voice.tools, voice.telephony · endpoint: `https://api.elevenlabs.io/v1/convai`

### 114. Retell AI API + MCP, B (69.4)

Hosted platform for phone and web voice agents, built as single-prompt agents or node-based conversation flows. Published per-minute price for every component, from $0.07 a minute all in. Call data kept indefinitely unless retention is set per agent.

- Page: https://www.anchorterminal.com/tools/retell-ai · Markdown: https://www.anchorterminal.com/tools/retell-ai.md · JSON: https://www.anchorterminal.com/api/v1/tools/retell-ai.json
- Capabilities: voice.agent, voice.pipeline, voice.speech-to-speech, voice.tools, voice.telephony · endpoint: `https://api.retellai.com`

### 126. Deepgram Voice Agent API, B (68.4)

One WebSocket that runs Deepgram STT (Flux or Nova-3), a managed or bring-your-own LLM and Deepgram or third-party TTS, with turn-taking, barge-in and function calling. $0.075 a minute all in on Standard, $0.050 with your own LLM and TTS, $200 of credit with no card. No phone numbers or SIP, so you bridge Twilio or another carrier yourself.

- Page: https://www.anchorterminal.com/tools/deepgram-voice-agent · Markdown: https://www.anchorterminal.com/tools/deepgram-voice-agent.md · JSON: https://www.anchorterminal.com/api/v1/tools/deepgram-voice-agent.json
- Capabilities: voice.agent, voice.pipeline, voice.tools · endpoint: `https://agent.deepgram.com/v1/agent`

### 191. Bland AI API + MCP, B (64.1)

Phone-first voice-agent platform that runs its own speech recognition, language model, TTS and telephony, billed as one per-minute rate. One per-minute rate covering STT, LLM, TTS and telephony, $0.14 on Start and $0.12 on Build. No OpenAPI document.

- Page: https://www.anchorterminal.com/tools/bland-ai · Markdown: https://www.anchorterminal.com/tools/bland-ai.md · JSON: https://www.anchorterminal.com/api/v1/tools/bland-ai.json
- Capabilities: voice.agent, voice.pipeline, voice.tools, voice.telephony · endpoint: `https://api.bland.ai/v1`

### 198. Vapi API + MCP, B (63.7)

Developer platform for phone and web voice agents. Voice agents can combine supported transcription, model and voice providers, including customer-supplied endpoints. Per-minute cost depends on the selected providers.

- Page: https://www.anchorterminal.com/tools/vapi · Markdown: https://www.anchorterminal.com/tools/vapi.md · JSON: https://www.anchorterminal.com/api/v1/tools/vapi.json
- Capabilities: voice.agent, voice.pipeline, voice.speech-to-speech, voice.tools, voice.telephony · endpoint: `https://api.vapi.ai`

### 279. Ultravox Realtime API, C (58.6)

Hosted voice agents on Ultravox, an open-weight model that takes speech directly with no speech-to-text step and answers through a TTS voice. Flat $0.05 a minute including model and built-in voices. No MCP server.

- Page: https://www.anchorterminal.com/tools/ultravox · Markdown: https://www.anchorterminal.com/tools/ultravox.md · JSON: https://www.anchorterminal.com/api/v1/tools/ultravox.json
- Capabilities: voice.agent, voice.speech-to-speech, voice.tools, voice.telephony · endpoint: `https://api.ultravox.ai/api`

### 296. Hume EVI (Empathic Voice Interface), C (57.3)

Hume's hosted voice-agent service, accessed through a WebSocket API. Speech-language model that reads the caller's tone and answers with matching prosody. One account-wide key, sent in the WebSocket and Twilio webhook URLs.

- Page: https://www.anchorterminal.com/tools/hume-evi · Markdown: https://www.anchorterminal.com/tools/hume-evi.md · JSON: https://www.anchorterminal.com/api/v1/tools/hume-evi.json
- Capabilities: voice.agent, voice.speech-to-speech, voice.pipeline, voice.tools, voice.telephony · endpoint: `https://api.hume.ai/v0/evi`

### 339. Bolna API + MCP, D (52.9)

Voice-agent platform built for Indian languages and phone campaigns. Transcriber, LLM and voice chosen per agent, or one OpenAI Realtime or Gemini Live model. Unsigned webhooks and tool calls, IP allowlisting only.

- Page: https://www.anchorterminal.com/tools/bolna · Markdown: https://www.anchorterminal.com/tools/bolna.md · JSON: https://www.anchorterminal.com/api/v1/tools/bolna.json
- Capabilities: voice.agent, voice.pipeline, voice.speech-to-speech, voice.tools, voice.telephony · endpoint: `https://api.bolna.ai`

### 352. Synthflow API + MCP, D (51.3)

Enterprise voice-agent platform built around a no-code Flow Designer and prompt builder, with a REST Platform API and a hosted MCP server. Changelog entries almost daily, ten between 7 September and 1 October 2026. Sales-led only, contracts from $30,000 a year, no free tier.

- Page: https://www.anchorterminal.com/tools/synthflow · Markdown: https://www.anchorterminal.com/tools/synthflow.md · JSON: https://www.anchorterminal.com/api/v1/tools/synthflow.json
- Capabilities: voice.agent, voice.pipeline, voice.tools, voice.telephony · endpoint: `https://api.synthflow.ai/v2`

### 383. Vogent API, D (47.4)

Phone voice-agent platform with its own conversational LLM, voices and IVR-navigation models alongside GPT. IVR detection and navigation models for outbound calls. No public changelog and no release found since January 2026.

- Page: https://www.anchorterminal.com/tools/vogent · Markdown: https://www.anchorterminal.com/tools/vogent.md · JSON: https://www.anchorterminal.com/api/v1/tools/vogent.json
- Capabilities: voice.agent, voice.pipeline, voice.tools, voice.telephony · endpoint: `https://api.vogent.ai/api`

## How we test this category

The same booking and support tasks on every platform. We measure response latency, how interruptions are handled, tool-call success, recovery after a misunderstanding and the total cost per completed call, and record whether the product runs a direct speech-to-speech model or an STT, model and TTS pipeline, since those are different setups. This test hasn't run yet, so Task success is pending and the grades here come from the categories assessed from public evidence.

## Indexed, not reviewed (1)

Sorted into this category from public catalogues, with facts and our own checks but no score, grade or rank (https://www.anchorterminal.com/indexed/index.md).

| Listing | Kind | What it does | Why it's here |
| --- | --- | --- | --- |
| [callwright](https://www.anchorterminal.com/tools/topness-callwright.md) | MCP server | Self-hosted MCP voice agent that places phone calls on your behalf via Retell. | vendor's own |