Conversational product teams
Developers adding a visual, speaking interface to an existing AI agent.
Independent tool overview
LiveAvatar is the production successor to HeyGen Labs' Interactive Avatar beta: a separate developer platform for embedding real-time, two-way AI avatar conversations in websites, apps, kiosks and agent experiences.
Visit the official LiveAvatar by HeyGen site ↗
Overview
HeyGen rebranded and moved its former Interactive Avatar beta to LiveAvatar, a separate platform with its own app, documentation, plans and credits. The product remains active, but new buyers should evaluate LiveAvatar rather than the old HeyGen Labs workflow.
LiveAvatar streams a speaking avatar in real time while a user talks or types. Full Mode packages speech recognition, an LLM, text-to-speech and avatar streaming; Avatar Only, also documented as LITE Mode, lets developers bring their own conversational stack and use LiveAvatar for the visual layer.
This is a developer product for customer support, tutoring, sales, recruiting, training and digital-assistant experiences—not HeyGen's separate tool for rendering downloadable marketing videos. The hardest work is usually the agent around the avatar: knowledge, safeguards, latency, escalation, consent and reliable handling of real user conversations.
Use cases
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Developers adding a visual, speaking interface to an existing AI agent.
Interactive tutors, coaches and role-play scenarios that benefit from face-to-face delivery.
Sales, support, recruiting, reception and kiosk experiences with clear escalation and disclosure.
Capabilities
Streams a 1080p preset avatar or eligible custom avatar during a live two-way conversation rather than generating a saved video.
LiveAvatar manages speech recognition, language-model orchestration, text-to-speech and the avatar session for the quickest setup.
Use your own LLM, speech and voice components while LiveAvatar supplies the real-time visual avatar layer.
Start with more than 100 presets or create a consented custom likeness from an image or recorded footage, subject to plan and approval requirements.
The product offers an API, web SDK and embeddable experience, with session tokens generated securely by a server.
Documentation covers conversational and push-to-talk modes, interruption handling, reconnect behavior and session transcripts in supported setups.
Current paid plans advertise unlimited concurrency, while session duration, resolution, credits and custom-avatar allowances vary by tier.
Process
Step 1
Use LiveAvatar for a real-time conversation; use HeyGen for prerecorded videos. Pick Full Mode for a managed voice stack or Avatar Only for more control.
Step 2
Build the interaction, disclosure, knowledge and escalation flow before investing in a custom likeness.
Step 3
Keep the API key on the server, issue short-lived session tokens and restrict what the agent can access or disclose.
Step 4
Measure first-frame and response latency, interruption behavior, background noise, failure recovery, factuality and human handoff.
Step 5
Estimate credits from session length and chosen mode, add overage and third-party AI costs, then monitor abandonment and useful task completion.
Cost
LiveAvatar subscriptions and credits are separate from HeyGen's video plans and HeyGen API wallet. Full Mode uses 2 credits per minute; Avatar Only / LITE Mode uses 1 credit per minute. The current public site advertises unlimited concurrency on paid plans, but maximum session duration and avatar resolution still vary.
$0
A small integration test with preset avatars.
$19 per month
A developer proof-of-concept plan.
$99 per month
The first self-serve tier with a branded custom avatar.
$475 per month
Higher-volume production use with a 1080p custom avatar.
Custom
Dedicated capacity and negotiated terms for large deployments.
Pricing checked . Check current pricing at the source ↗
Assessment
Compare
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Content Creator
Choose Tavus when conversational video interfaces and digital-replica APIs are the main comparison.
Explore tavus →Marketing
Choose D-ID for talking-avatar and visual-agent workflows with a different creation and integration stack.
Explore D-ID →Marketing
Choose Synthesia when the goal is polished prerecorded training or business video rather than a live two-way avatar.
Explore Synthesia →Questions
Yes, under a new product. LiveAvatar is the official production successor to the HeyGen Labs Interactive Avatar beta and now runs on a separate LiveAvatar platform.
HeyGen creates prerecorded, downloadable AI videos. LiveAvatar streams a real-time two-way avatar conversation through an API, SDK or embed. Their plans, workspaces and credits are separate.
Full Mode includes speech recognition, LLM, voice and avatar streaming and uses 2 credits per minute. Avatar Only, also called LITE Mode in the documentation, lets you provide the conversational stack and uses 1 credit per minute.
HeyGen says existing Interactive Avatar users can receive migration support. The old and new avatar systems are not directly cross-compatible, so contact LiveAvatar support for the specific asset.
Its primary output is a live, ephemeral WebRTC stream, not a permanent MP4. If the use case needs a reusable rendered video, the separate HeyGen product is a better fit.
Divide credits by two for approximate Full Mode minutes, or use one credit per minute in Avatar Only / LITE Mode. Actual billing follows session time and any enabled overage.
Bottom line
LiveAvatar is the right continuation of HeyGen's Interactive Avatar for teams building a real-time visual agent. It is most compelling when the face materially improves an already useful conversational workflow. Prototype with a preset, validate task completion and latency first, then choose Full or Avatar Only based on control, cost and the strength of your existing voice stack.
Visit LiveAvatar by HeyGen website ↗
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.