Prompt

A prompt is the instruction and context given to a language model for a specific response, including the user's message, system rules, retrieved content, and tool results when available.

How prompts work

A prompt is the instruction and context given to a language model for a specific turn or task. In an avatar system, the prompt helps decide what the avatar says, what tone it uses and what information it should rely on.

The prompt can include more than the user's latest message. It may also include the system prompt, conversation history, retrieved documents, tool results and product-specific rules.

A concrete example: a user asks an onboarding avatar how to connect a webhook. The prompt may include the user's question, the relevant docs snippet and an instruction to answer step by step.

Prompt quality matters because the avatar speaks the model's output directly. Loose prompts lead to vague or risky answers; specific prompts make the avatar more useful and predictable.

What Anam ships

Anam's Cara-4 model delivers expressive real-time avatars with around 150 ms server-side avatar-generation latency once a session is running, across 70+ languages. Builders use JavaScript and Python SDKs or integrations for LiveKit, Pipecat, ElevenLabs Agents, Agora, and VideoSDK. Bring any AI stack including OpenAI, Claude, Gemini, Mistral, Groq, Deepgram, Cartesia, or custom providers. The platform supports WebRTC delivery, SOC 2 Type II, HIPAA, zero data retention, and regional data residency. Sessions stream low-latency audio and video to browsers and native apps.

Explore the real-time avatar API →

Related terms

System prompt

A system prompt is the persistent base instruction that sets an avatar's role, tone, boundaries, allowed behaviours, and escalation rules across the whole conversation.

Persona

A persona is the configured identity of an avatar, including its role, tone of voice, appearance, behaviour, knowledge boundaries, and rules for how it should respond to users.

Grounding

Grounding is the practice of anchoring a model's answer to trusted source content, retrieved documents, or tool results so an avatar can respond with verified information instead of guessing.

Hallucination

A hallucination is when a language model gives a confident but incorrect or unsupported answer, which is especially risky in avatar experiences because the response is delivered through a persuasive face and voice.

Frequently asked questions

What does a prompt do in an avatar conversation?

A prompt gives the model instructions, context, and the user's current turn, so it can decide what the avatar should say or do next.

Why are prompts important for real-time avatars?

The avatar speaks the model's output out loud, often immediately. A prompt has to guide behavior clearly enough that the response is useful, safe, and natural on the first attempt.

How is a prompt different from a system prompt?

A system prompt sets durable rules for the agent. A prompt can also include the current user message, retrieved context, temporary task instructions, or tool results for one turn.

What makes a good avatar prompt?

It is concise, grounded, and specific about role, tone, task, limits, and response style. For avatars, it should also produce speech that sounds natural when spoken aloud.

Last updated: 17th July 2026 · Reviewed quarterly.

← Back to glossary