Convai Unreal Engine plugin

Find installation guides, feature documentation, Blueprint reference, and troubleshooting pages for the Convai Unreal Engine plugin.

The Convai Unreal Engine plugin connects Unreal Engine 5 projects to Convai, bringing conversational AI into training simulations, interactive experiences, and games. Actors in a scene hold real-time voice and text conversations, express emotions, animate faces in sync with speech, and respond to player actions — all configurable per project. Core runtime workflows are Blueprint-first: configure characters, drive conversations, and wire most feature behavior from Blueprint graphs without writing C++. Each feature is opt-in and builds on the core components; add only what your project needs.

New to the plugin? Follow Getting started to install, configure your API key, and run your first AI character.

Learn the plugin

Overview What the plugin is, how its modules are structured, and a map of every feature.

Compatibility and requirements UE 5.x version support, platform targets, and character rig compatibility.

Getting started Install the plugin, configure your API key, and add your first AI character.

Core concepts Runtime architecture, session lifecycle, conversation flow, and Blueprint events.

Features

Each feature is opt-in and builds on the core components.

Character actions Characters execute typed in-scene commands (Move To, Follow, custom).

Dynamic context Push live world state and events into character knowledge at runtime.

Emotion Convai infers emotion from conversation and drives blendshape expressions.

Long-term memory Characters remember each player across sessions using an end-user ID.

Narrative design Trigger scripted conversation branches and sections by name from Blueprint.

Scene metadata Tag level actors so characters are aware of and can act on them.

Vision Stream camera frames to Convai so characters can describe what they see.

Gaze attention Route the object under the player's crosshair as context to the active character.

Lip sync Drives mouth and facial blendshapes from response audio in real time.

Reference

Advanced topics Production authentication and integration guides beyond the core setup path.

Blueprint reference Full property, function, and event surface for all components.

The Convai editor window Sign in, browse samples, and export diagnostic logs from within the editor.

Troubleshooting Common failure modes, diagnostic steps, and known issues.

Latest release

4.0.0-beta.21 includes gesture and pointing animations triggered automatically by LLM-issued character actions, UConvaiObjectComponent for tagging level actors with live scene properties, gaze-driven attention with silhouette highlight and on-screen cursor, and fuzzy enum matching so minor AI spelling variations no longer drop action parameters. See Release notes for the full changelog.

Next steps

Getting started