Free
$0
Developer tier for testing CVI
- 20 CVI minutes/month
- 1 concurrent session
- 5 min max conversation
- 10 stock faces

Real-time conversational AI video agents with lifelike faces, voice and memory, built via API
Tavus is a San Francisco AI research lab that builds real-time, face-to-face AI agents it calls PALs. A PAL can see, hear and respond in a live video conversation, with a configurable persona, rules and persistent memory across sessions. Developers can build these experiences through the Conversational Video Interface (CVI) API, while non-developers can create a PAL with the no-code PAL Maker. Tavus says its platform is used by more than 150,000 developers.
Plans as listed by the vendor on . Check their site for current prices.
$0
Developer tier for testing CVI
$22/month
Entry paid developer plan
$59/month
Small teams scaling usage
$397/month
Production workloads
$975/month
High-volume deployments
Custom/custom
Managed, white-labeled deployments
Scored out of 10 by the Finder Atlas editorial team.
Other tools people compare with Tavus.
Generate and edit video and images from text or images with Luma's Ray models
Describe the edit in plain English and an AI agent builds it on an editable timeline
AI video editor that turns raw footage and prompts into finished, captioned videos
AI video, image and audio creation suite with Gen-4.5, agents and an API
Generate AI video and images from text, photos and references in one studio
No reviews yet
Used Tavus? Help others decide by sharing your experience.
A PAL is Tavus's term for a real-time AI agent that appears as a video persona, can see and hear the user, and keeps memory across conversations.
Plans include a monthly allotment of CVI minutes. Most paid plans allow pay-as-you-go overage, billed per minute rounded to 6 seconds with a 30-second minimum per conversation.
Yes. Tavus supports its built-in LLM or a bring-your-own LLM setup.
Yes. Tavus lists Zoom, Google Meet and Microsoft Teams as supported call platforms.
Yes. The developer Free plan includes 20 CVI minutes per month with one concurrent session and a 5-minute conversation limit.
Image & Visual AI
AI video, image and audio creation suite with Gen-4.5, agents and an API
Video AI
Edit video and podcasts by editing their transcript, with AI audio cleanup and clips
Video AI
Create avatar videos from text and translate them into 175+ languages