> ## Documentation Index
> Fetch the complete documentation index at: https://docs.trugen.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Huma-1

> TruGen's first avatar model to combine full-face rendering with real-time emotional control — powered by Gaussian Avatars.

**Huma-1 is TruGen's first-generation avatar renderer** — the first model to combine full-face rendering with real-time emotional control. Powered by **Gaussian Avatars**, it captures every micro-expression, lip movement, blink, and emotional nuance to deliver hyper-realistic digital humans that don't just talk, but truly communicate.

<div style={{ overflow: "hidden", borderRadius: "12px", lineHeight: 0, marginLeft: "-1.5rem", marginRight: "-1.5rem", marginTop: "1rem", marginBottom: "1rem" }}>
  <img src="https://mintcdn.com/trugen/9qi__syJVXNUbiUr/images/huma-hawkeye.png?fit=max&auto=format&n=9qi__syJVXNUbiUr&q=85&s=247bdeea6026a71df25842087db6aac4" alt="Huma-1 and Huma-2 avatar models with Hawkeye-1 vision model" style={{ width: "100%", display: "block" }} width="647" height="325" data-path="images/huma-hawkeye.png" />
</div>

## Lightning-fast rendering

Huma-1 renders the full face and expressions in **under 80 milliseconds** — fast enough to power seamless, two-way, real-time conversations without perceptible lag. Whether it's a customer support interaction or a coaching session, the experience feels natural and immediate.

## Emotion-aware expression engine

Most avatar systems treat facial animation as a surface-level overlay. Huma-1 goes deeper. Its emotion-aware engine adapts expressions, tone, and micro-gestures based on conversational flow — meaning the avatar doesn't just match words to mouth shapes; it mirrors the emotional arc of the dialogue. The result is an avatar that feels genuinely responsive and alive.

## Full-face animation

Huma-1 tackles the four pillars of believable facial animation:

* **Micro-expressions** — subtle eyebrow shifts, gentle smiles, and involuntary cues that make faces feel alive
* **Lip syncing** — perfect word-to-lip synchronisation in real time, making dialogue feel natural and fluid
* **Eye blinking** — natural blink patterns that prevent the dead-stare effect plaguing most avatar systems
* **Emotional presence** — avatars adjust reactions to reflect warmth, concern, excitement, or empathy in context

## Better together: Huma-1 + Hawkeye-1

Huma-1 doesn't work in isolation. Paired with [Hawkeye-1](/docs/agents/vision/overview) — our vision model — it gains the ability to understand context, emotion, and nuance from the person on the other side of the conversation. Hawkeye analyses tone, pacing, and intent, while Huma mirrors it with human-like timing and context-aware engagement. Together, they power conversational video agents that don't just respond — they connect.

## Key characteristics

| Attribute           | Detail                                      |
| ------------------- | ------------------------------------------- |
| Modality            | Real-time video generation                  |
| Speech sync latency | ≤ 80 ms                                     |
| Rendering approach  | Gaussian Avatars                            |
| Expression fidelity | Micro-expressions, gaze, head motion, blink |
| Output              | Streamed video frames                       |

## Next steps

<CardGroup cols={2}>
  <Card title="Models overview" icon="brain" href="/docs/models/overview">
    Compare Huma-1 against Huma-2.
  </Card>

  <Card title="Huma-2" icon="bolt" href="/docs/models/huma-2">
    The current default real-time renderer.
  </Card>
</CardGroup>