Provider Feature Comparison
What each engine supports in VoiceAIWrapper, side by side: VAW Native, Vapi, Retell AI, ElevenLabs, and the legacy Bolna and Ultravox Pods.
Written By voiceaiwrapper
Last updated 12 days ago
Each Pod (an engine or provider connection) runs on one engine, and the engine decides what you can do with the agents on it. Use this page to pick the right engine for each client: for example, VAW Native when you want to build and bill everything inside VoiceAIWrapper, or a client's own Vapi account when they already have agents there.
The tables cover what works inside VoiceAIWrapper. Your provider may offer more in its own dashboard.
Bolna and Ultravox are legacy engines. You can't add new Bolna or Ultravox Pods, but existing ones keep working. See Bolna and Ultravox Pods (Legacy).
The engines in short
- VAW Native is VoiceAIWrapper's built-in voice engine. Every organization has one VAW Native Pod. Calls are paid with AI credits, and you build agents, knowledge bases and tools inside VoiceAIWrapper.
- Vapi, Retell AI and ElevenLabs are BYOK Pods: you connect your own provider account with its API keys. You build agents in the provider's dashboard (or with VAW Assistant), and the provider bills you for calls.
Calls and channels
The website widget needs a plan that includes Web widget.
Chat, pre-call webhooks and voicemail
The inbound pre-call webhook lets your server pass caller details to the agent, or turn the caller away, before the agent answers. Every engine supports the outbound pre-call webhook.
Phone numbers
Phone number pools need a plan that includes Phone number pools. Vapi and Retell AI numbers must be synced before a campaign can use them. See Sync Webhooks So VoiceAIWrapper Can Track Calls.
Building agents
On Vapi, Retell AI and ElevenLabs, VAW Assistant (the guided builder that creates an agent in your provider account) needs a plan that includes VAW Assistant. Call Data Extractions are available on VAW Native only.
Clients and billing
You decide what each client can see and change in their workspace. See Set Client Permissions on VAW Native Workspaces and Set Client Permissions on Vapi, Retell and ElevenLabs Workspaces.
Which engine should I use?
- Choose VAW Native if you want one bill (AI credits), want to build everything in VoiceAIWrapper, or want clients to edit more than the prompt.
- Choose Vapi or Retell AI if you or your client already have agents there. Vapi workflows and squads also work in campaigns.
- Choose ElevenLabs if you build your agents in ElevenLabs. Remember that ElevenLabs has no inbound pre-call webhook in VoiceAIWrapper.
You don't have to pick one. You can connect several Pods and use a different engine per client. See Use Several Providers and Organizations in One Account.
Related articles
Was this helpful?
More in VAW Native vs Vapi, Retell AI and ElevenLabs
Setup and Upkeep: No Provider Keys or Webhook SyncingBilling: One Balance in AI Credits Instead of a Separate Provider BillFeatures You Only Get on VAW NativeClient Controls: What Your Clients Can Do on Each EngineStill need help? Share an idea