Avaluma Agents: Create and Configure Avatars in the Dashboard

Avaluma gives you two ways to run a photorealistic conversational avatar. The fastest is to create an Agent in the Avaluma dashboard at avaluma.ai: you configure the avatar, voice, and behavior in your browser, and Avaluma hosts everything for you. If you need full control over compute and data, you can instead run the pipeline yourself — see Self-Hosting.

Agents (hosted)

Create an agent in the dashboard, share a link, or embed it. Avaluma runs the avatar server and the voice pipeline for you.

Self-Hosting

Run the Avatar Server and LiveKit Agent on your own hardware for full control over data, models, and compute.

What is an Agent?

An Agent is a configured, hosted avatar that lives in your Avaluma account. Each agent bundles everything a live conversation needs:

A photorealistic avatar rendered from an Avaluma .hvia model
A voice for text-to-speech
An AI persona — the system prompt and behavior that shape its responses
Sharing settings — a public link and optional embedding on your own website

Once configured, an agent has a stable Agent ID (a UUID) that you use to share or embed it.

Create an agent

Create a new agent

Click New agent, give it a name, and select the avatar you want it to use. Your available avatars come from the .hvia files issued with your license.

Configure voice and persona

Choose a voice for text-to-speech and write the agent’s instructions (system prompt). The instructions define the agent’s tone, knowledge, and behavior during a conversation. A default voice is already preselected — a voice must be selected for the agent to speak.

Save and test

Save the agent and start a test conversation from the dashboard to confirm the avatar speaks and responds as expected.

Configure an agent

Open an agent to adjust its configuration at any time. Common settings include:

Setting	Description
Name	Internal label for the agent in your dashboard.
Avatar	The `.hvia` avatar the agent renders.
Voice	The text-to-speech voice used for spoken replies.
Model	The ElevenLabs speech model used to generate the voice — see Choosing a speech model.
Instructions	The system prompt that shapes the agent’s persona and behavior.
Language	Default conversation and UI language.
Embedding	Whether the agent may be embedded via `<iframe>`, and on which origins.
Kiosk	Standalone fullscreen page for displays (kiosk). Available for every agent; you find the URL and QR code in the Deployment tab.
Web search	Whether the agent may look things up on the internet during a conversation — see Web search.

The direct link (/agent/AGENT_ID) and the kiosk are reachable for every agent, so treat the Agent ID like a public link. Only embedding via iframe needs to be enabled separately — together with the origins it is allowed on.

Choosing a speech model

Under Voice you can also pick the ElevenLabs model that generates the speech. Every model offered works with your own instant voice clones (voice ID).

Model	Latency	Languages	When to use
v4 Turbo (default)	~100 ms	90+	Preselected for new agents — highly expressive and fast.
Flash v2.5	~75 ms	32	Lowest latency when expressiveness is secondary.
v3 Conversational	~280 ms	70+	Expressive, but slower than v4 Turbo.
v4	~550 ms	90+	For comparison only — too slow for live conversations.
Flash v2	~75 ms	English only	English-only agents.
Multilingual v2	higher	29	High audio quality when latency is secondary.
Turbo v2.5	~250 ms	32	The model of older agents. Same voice as Flash v2.5, only slower; ElevenLabs recommends switching.

Voice settings

Under Voice Settings you adjust Stability, Similarity and Speed with sliders. Each slider has a Voice default checkbox above it. While it is checked, the value is not sent and ElevenLabs uses the voice’s own setting. Uncheck it to enable the slider; it then starts at the ElevenLabs defaults: stability 0.5, similarity 0.75, speed 1.0. Which sliders appear depends on the model:

Model	Stability	Similarity	Speed
v4 Turbo, v4	continuous	continuous	—
v3 Conversational	only `0` (Creative), `0.5` (Natural), `1` (Robust)	—	—
all others	continuous	continuous	0.7–1.2

The sliders adapt automatically once you switch the model: unsupported ones disappear, and on v3 a stability value in between is rounded to the nearest preset.

New agents start on v4 Turbo. Existing agents keep their model until you change it yourself.

Defaults for new agents

A newly created agent starts with this configuration. You can change all of it in the form afterwards:

Area	Default
Speech recognition (STT)	Deepgram Flux `flux-general-multi`, language English
Language model (LLM)	LiveKit Inference `gemma-4-31b-it`
Speech output (TTS)	ElevenLabs v4 Turbo
Turn-taking	STT endpointing (Deepgram Flux), endpointing dynamic (adapts to speech pauses), preemptive TTS on (speech output starts before the end of turn)

If you switch speech recognition away from Deepgram Flux, turn-taking is reset to “Not configured”, because STT endpointing only works with Flux.

Web search

With web search, the agent can look up current information on the internet during a conversation — for example opening hours, news, or prices that are not in its background knowledge. Turn it on in the Background knowledge tab under Tools with the Web search checkbox, then save the agent. It is off by default. Web search works with all language models except OpenAI Realtime. With the default model gemma-4-31b-it and with Cerebras, GPT-5.4 mini runs the search in the background; a search then takes a few seconds before the agent answers.

Manage agents

In the Agents overview, each agent has a menu (three dots) with additional actions:

Action	Description
Duplicate	Creates a full copy of the agent — including voice, instructions, tools, and all uploaded documents. The copy gets a new Agent ID and the name `<Name> (Copy)`.
Transfer	Hands the agent over to another user (by email address). The new owner gets full access, and the agent disappears from your list.
Delete	Permanently removes the agent. This action cannot be undone.

Duplicating also carries over the embedding settings (including the allowed origins). The copy is therefore immediately reachable under the same conditions as the original.

Per-agent usage statistics

In the Agents overview, each agent card shows how many conversation minutes that agent has been used in the current month. The statistics icon on the card opens a window with the key metrics:

Metric	Description
Conversations	Number of conversations in the selected period.
Minutes	Total conversation minutes used.
Avg. length	Average length of a conversation.
Last used	Date of the most recent conversation.

A trend additionally shows conversations over time, switchable between 1 day (by hour), 7 days, and 30 days. Use the date field to pick the end date of the period (default: today) and look back at past days or weeks.

The monthly total across all agents and your plan limit are shown in the bar at the top of the Agents page — see Pricing & Billing for details.

Each agent is reachable at a hosted URL based on its Agent ID:

https://avaluma.ai/agent/AGENT_ID

To put the agent directly on your own site, embed it as an <iframe> and control it from your page.

Embedding

Embed an agent on your website with an <iframe> and customize its appearance and behavior via query parameters.

Messaging

Control the embedded avatar programmatically and receive events with the postMessage API.

Kiosk mode

Run the agent fullscreen on an unattended display — with a screen wake lock and idle auto-reset.

Agents (hosted)

Self-Hosting

​What is an Agent?

​Create an agent

​Configure an agent

​Choosing a speech model

​Voice settings

​Defaults for new agents

​Web search

​Manage agents

​Per-agent usage statistics

​Share and embed

Embedding

Messaging

Kiosk mode

What is an Agent?

Create an agent

Configure an agent

Choosing a speech model

Voice settings

Defaults for new agents

Web search

Manage agents

Per-agent usage statistics

Share and embed