The reliability in production. We build outbound voice AI agents for clients and Retell is the first platform we've trusted enough to put in front of real leads at scale. Calls handle interruptions naturally, recover from unexpected responses without breaking, and the post-call analysis returns structured data (booked status, objections, extracted variables) that hits the CRM automatically via webhook. The ElevenLabs voice integration is seamless and the whole stack, from LLM config to live calls, took less than a day to deploy. Review collected by and hosted on G2.com.

The documentation could go deeper on edge cases and the dashboard analytics could be more prominent. Call performance data is there but you have to dig for it. Neither is a big deal, just areas with room to grow. Review collected by and hosted on G2.com.

What I like most about Retell AI is its flexibility for building serious voice agents beyond simple prompt-based bots. The platform gives good control over call behavior, speech settings, custom LLM integrations, and structured conversation flows. I also like that it supports advanced use cases like multilingual assistants, external tools, and custom backend logic, which makes it useful for real production scenarios rather than only demos. The interface is modern, testing is convenient, and it is clear that the product is built for developers and teams who want more control over the voice stack. Review collected by and hosted on G2.com.

The biggest downside is that setup can become confusing once you move beyond the basic use case. Some configuration areas, especially around custom LLMs, voices, transcription, and function/tool behavior, are not always intuitive, and it can take time to understand how everything fits together. There is a noticeable learning curve for advanced integrations, and debugging complex voice flows or backend-connected agents can be harder than it should be. I would also like clearer documentation and more step-by-step guidance for production-style setups, especially when combining multilingual transcription, custom backends, and external voice providers. Review collected by and hosted on G2.com.

I’ve been using Retell AI to build voice agents, and the experience has been phenomenal. The 600ms latency makes conversations feel incredibly natural. The API is very well-documented, making it easy to integrate with CRMs and other tools. It’s a game-changer for anyone looking to automate phone calls with high quality and reliability. Review collected by and hosted on G2.com.

While Retell AI has the best latency in the market, the pricing model is a bit steep for small startups. Also, the initial setup and prompt engineering require a steep learning curve. I wish there were more pre-built templates for common use cases to make the onboarding process faster. Review collected by and hosted on G2.com.

What I like most about Retell AI is how easy it is to build production-ready voice AI agents that can handle real phone calls. Review collected by and hosted on G2.com.

One area that could improve is providing more built-in templates and guided setup for common business use cases like real estate, healthcare, and customer support. While the platform is powerful, new users may need some time to understand the best way to structure prompts, call flows, and webhook integrations.

It would also be helpful to have more native integrations with popular CRMs and automation tools. Currently these integrations are possible through webhooks, which works well, but direct integrations could make onboarding even easier for non-technical users.

Overall these are minor improvements and the platform is already very capable for building voice AI solutions. Review collected by and hosted on G2.com.

Retell lets you create any voice agent you can imagine, powering it up with super powers, using MCP tools and functions. The response time and the top tier quality of the voice is something that will keep blowing people's minds every single time they interact with one of these agents. Review collected by and hosted on G2.com.

Nothing. I am thrived to see how this company scales in the long run. It will definetely impact the whole world, and we are starting to see its true power. Review collected by and hosted on G2.com.

Not having to deal with delicate pipelines Review collected by and hosted on G2.com.

I had 1 persistent error that took days to fix Review collected by and hosted on G2.com.

“I really like the flexibility Retell AI offers. It’s easy to customize voice agents, integrate APIs, and quickly iterate on call flows without a lot of overhead. Review collected by and hosted on G2.com.

One thing I found a bit challenging is adding phone numbers via SIP trunking. The setup process feels more complex than it should be, especially compared to platforms like Vapi where it’s much more straightforward. It would be great if this part was simplified to make onboarding easier. Review collected by and hosted on G2.com.

I was looking for a quick and easy way to make a prototype for C-level executives and Retell AI was the best out there. Easy to use, professional, and exceptional. Review collected by and hosted on G2.com.

No connection with WhatsApp and limited language support. Review collected by and hosted on G2.com.

What I like best about Retell AI is how fast and easy it is to build realistic AI voice agents that sound natural and can handle real business conversations. It offers powerful integrations, smooth conversational flow, and enough flexibility to create customised solutions for industries like healthcare, real estate, legal, and customer support without needing to build everything from scratch. Review collected by and hosted on G2.com.

One thing I dislike about Retell AI is that some advanced customisations and integrations still require external tools and technical setup, especially for complex workflows. Pricing can also become expensive at scale, and support for international business numbers and region-specific telephony options could be improved further. Review collected by and hosted on G2.com.

What I like best about Retell AI is how quickly you can go from an idea to a fully working AI phone agent that actually sounds natural on real calls. The platform makes it easy to build AI receptionists that can answer questions, qualify leads, and book appointments without needing a huge engineering setup. I also like how flexible it is with integrations, so you can connect it to CRMs, booking systems, and other tools businesses already use. That makes it practical for real deployments instead of just demos. Overall, it feels built for people who want to ship AI voice agents that businesses can actually rely on. Review collected by and hosted on G2.com.

One downside I’ve noticed with Retell AI is that the native scheduling integration is mainly built around Cal.com. While Cal.com works well, many businesses already use other booking platforms, so it would be great if there were more built-in integrations for different scheduling tools to make it easier to fit into existing workflows. Review collected by and hosted on G2.com.