# Tavus
ID: 08-tavus
Page: https://gtmmcps.com/tools/08-tavus
JSON: https://gtmmcps.com/data/tools/08-tavus.json
Documentation references: https://gtmmcps.com/docs/08-tavus/
Category: Video Prospecting
MCP support: official
Testing: RESEARCHED
API access: free
Last checked: 2026-08-24
Endpoint check: live-auth-gated; date: 2026-09-04

Developer platform for building real-time, two-way conversational AI video agents ("Conversational Video Interface") - positioned for GTM use cases like greeting website visitors and booking meetings, not primarily batch personalized-outbound video.

## MCP links
- https://mcp.tavus.io/mcp
- https://docs.tavus.io/sections/agent-tools/mcp-server.md
- https://docs.tavus.io/sections/agent-tools/mcp-tools-reference.md
- https://github.com/rakeshdavid/Tavus-MCP

## Sources
- https://www.tavus.io/
- https://www.tavus.io/pricing
- https://docs.tavus.io/llms.txt
- https://docs.tavus.io/sections/agent-tools/mcp-server.md
- https://github.com/rakeshdavid/Tavus-MCP
- https://coldiq.com/blog/tavus-video-prospecting

## Research notes
Third-party blogs (e.g. ColdIQ) frame Tavus as a "video prospecting" tool via its replica/clone capability, but Tavus's own current marketing emphasizes real-time conversational agents over the "record once, blast personalized clips" model most other entries in this file use - worth noting the mechanism differs from Vidyard/Sendspark/Potion even though the sales use case is real.

## Named capabilities
Tool names below are reported by sources. Their presence is not a successful call.
- create_conversation: Create a new conversational video interface; evidence: source; party: third-party
- create_lipsync: Create a lipsync video by synchronizing audio with video; evidence: source; party: third-party
- create_persona: Create a new persona for conversational AI; evidence: source; party: third-party
- create_replica: Create a new AI replica from a training video; evidence: source; party: third-party
- delete_conversation: Delete a conversation permanently; evidence: source; party: third-party
- delete_lipsync: Delete a lipsync permanently; evidence: source; party: third-party
- delete_persona: Delete a persona permanently; evidence: source; party: third-party
- delete_replica: Delete a replica permanently; evidence: source; party: third-party
- delete_speech: Delete a speech permanently; evidence: source; party: third-party
- delete_video: Delete a video permanently; evidence: source; party: third-party
- end_conversation: End an active conversation; evidence: source; party: third-party
- generate_speech: Generate speech audio from text using a replica; evidence: source; party: third-party
- generate_video: Generate a video using a replica and script or audio; evidence: source; party: third-party
- get_conversation: Get details of a specific conversation; evidence: source; party: third-party
- get_lipsync: Get details of a specific lipsync; evidence: source; party: third-party
- get_persona: Get details of a specific persona; evidence: source; party: third-party
- get_replica: Get details of a specific replica; evidence: source; party: third-party
- get_speech: Get details of a specific speech; evidence: source; party: third-party
- get_video: Get details of a specific video; evidence: source; party: third-party
- list_conversations: List all conversations in your account; evidence: source; party: third-party
- list_lipsyncs: List all lipsyncs in your account; evidence: source; party: third-party
- list_personas: List all personas in your account; evidence: source; party: third-party
- list_replicas: List all replicas in your account; evidence: source; party: third-party
- list_speeches: List all speeches in your account; evidence: source; party: third-party
- list_videos: List all videos in your account; evidence: source; party: third-party
- patch_persona: Update a persona using JSON patch format; evidence: source; party: third-party
- rename_replica: Rename an existing replica; evidence: source; party: third-party
- rename_speech: Rename an existing speech; evidence: source; party: third-party
- rename_video: Rename an existing video; evidence: source; party: third-party

## CLI evidence
Status: community
- Reported install command: pip install tavus-cli; source: https://pypi.org/project/tavus-cli/
- Reported install command: pip install tavus; source: https://pypi.org/project/tavus/

Read current vendor documentation before configuring or running commands.