# Best conversational voice-agent APIs (slim)

> ElevenLabs Agents API + MCP (BB), Retell AI API + MCP (B) and Deepgram Voice Agent API (B) lead the 14 ranked conversational voice-agent APIs. Picks by need, strengths, weaknesses and prices from the Anchor benchmark.

- Full: https://www.anchorterminal.com/best/voice-agents/index.md (~5,550 tokens) · this version ~1,430 tokens · JSON https://www.anchorterminal.com/best/voice-agents/index.json · canonical https://www.anchorterminal.com/best/voice-agents/
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-09

The 10 highest-scoring of 14 conversational voice-agent APIs on the Anchor benchmark, with a pick for each need and where each one falls short. Scores come from public evidence, re-checked as vendors change.

- Ranked: 14 · agent-ready (BB or better): 1 · accept x402: 0 · hosted endpoints: 14
- Full ranked table: https://www.anchorterminal.com/categories/voice-agents.md
- Head-to-head comparisons: https://www.anchorterminal.com/compare/voice-agents/index.md (91)
- Methodology: https://www.anchorterminal.com/benchmark/index.md

## The shortlist

| # | Tool | Grade | Score | Best for | Price | Where |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [ElevenLabs Agents API + MCP](https://www.anchorterminal.com/tools/elevenlabs-agents.md) | BB | 71.3 | Teams that want a managed pipeline with ElevenLabs voices, a choice of LLM, wide telephony integration and web and mobile SDKs. | $22 / mo | hosted |
| 2 | [Retell AI API + MCP](https://www.anchorterminal.com/tools/retell-ai.md) | B | 69.1 | Teams that want a self-serve phone agent with clear per-component pricing, scoped keys and a choice of LLM and voice from Retell's menu. | $2 / mo | hosted |
| 3 | [Deepgram Voice Agent API](https://www.anchorterminal.com/tools/deepgram-voice-agent.md) | B | 68.1 | Developers who already run telephony (Twilio, Amazon Connect, Genesys, AudioCodes) and want a cheap managed pipeline with their choice of LLM. | Pay per use | hosted and local |
| 4 | [AssemblyAI Voice Agent API](https://www.anchorterminal.com/tools/assemblyai-voice-agent.md) | B | 64.4 | A developer who wants one managed pipeline at a flat rate and is content with AssemblyAI's own speech models and voices, with an optional model of their own. | Pay per use | hosted |
| 5 | [Bland AI API + MCP](https://www.anchorterminal.com/tools/bland-ai.md) | B | 63.8 | Teams that want one predictable per-minute bill for outbound and inbound phone agents, and personal agents that need their own US number. | $299 / mo | hosted and local |
| 6 | [Vapi API + MCP](https://www.anchorterminal.com/tools/vapi.md) | B | 63.5 | Developers who want to choose every provider, bring their own keys and tune cost against latency. | $29 / mo | hosted and local |
| 7 | [OpenAI Realtime API](https://www.anchorterminal.com/tools/openai-realtime.md) | B | 63.3 | Teams building their own voice agent on one speech-to-speech model who want browser, server and phone transports from one vendor and can run a backend for secrets and tools. | Pay per use | hosted |
| 8 | [Ultravox Realtime API](https://www.anchorterminal.com/tools/ultravox.md) | C | 58.5 | Cost-sensitive voice agents that bring their own telephony and want a speech-native model. | $100 / mo | hosted |
| 9 | [Gemini Live API](https://www.anchorterminal.com/tools/gemini-live.md) | C | 57.1 | Teams building their own voice or vision assistant on a single speech-to-speech model with function calling and Search grounding, who can run a backend for tokens and audio transport. | $14 / 1k req | hosted |
| 10 | [Hume EVI (Empathic Voice Interface)](https://www.anchorterminal.com/tools/hume-evi.md) | C | 56.7 | Consumer and coaching products where the caller's tone matters and English is enough on EVI 3. | $14 / mo | hosted |

## Picks by need

- Highest score overall: [ElevenLabs Agents API + MCP](https://www.anchorterminal.com/tools/elevenlabs-agents.md), BB, 71.3/100 on the benchmark. Also [Retell AI API + MCP](https://www.anchorterminal.com/tools/retell-ai.md), B, 69.1/100.
- Reliability: [Vapi API + MCP](https://www.anchorterminal.com/tools/vapi.md), 70/100 on reliability, against 60 for the overall leader.
- Agent ergonomics: [Vocily AI](https://www.anchorterminal.com/tools/vocily.md), 84/100 on agent ergonomics, against 77 for the overall leader.
- Maintenance & community: [Vapi API + MCP](https://www.anchorterminal.com/tools/vapi.md), 88/100 on maintenance & community, against 85 for the overall leader.
- Self-hosting under an open licence: [Bolna API + MCP](https://www.anchorterminal.com/tools/bolna.md), self-hosted, MIT licence.

## How to choose

- Speech-to-speech or pipeline: Check whether the product runs one speech-to-speech model or a separate recognition, language and synthesis chain, since each changes latency and what you can swap.
- Handling of interruptions: Ask how the platform stops speaking when a caller interrupts, because a slow stop leaves the agent talking over the customer.
- Tool calls during a live call: Check how a tool call is confirmed to the caller and what happens when it fails, since a long pause or a wrong booking is what the caller hears.
- Cost per completed call: Work out the cost of a finished call from per-minute, language model and telephony charges together, because a low per-minute rate can hide a costly language model.

- How the benchmark tests 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.

Each listing's verdict, strengths and weaknesses: https://www.anchorterminal.com/best/voice-agents/index.md