做数字人、虚拟角色或交互式 AI 的团队终于有了一个统一框架——Archon 把文本、音频、动作、视频全打通了，不用再拼凑多个模型，做沉浸式体验的开发者可以直接参考其架构。
浙江大学团队提出 Archon，一个完全预训练的统一多模态模型，用于生成包含文本、音频、动作和视觉内容的完整数字人。该模型通过模态专用分词器和原生自回归架构，统一了七种模态，并在 72 个任务上预训练以建模联合分布。为解决高保真对话视频中的 token 爆炸问题，Archon 引入了一种内存高效的语义视频重参数化方法，实现 4 倍 token 压缩同时保留精细动态，并配合语义驱动的视频扩散解码器。此外，提出的“模态思考”机制将模糊的跨模态任务分解为逐步推理，提升了生成保真度和可控性。实验表明，Archon 在多种数字人生成任务上达到或超越现有水平。
Archon: A Unified Multimodal Model for Holistic Digital Human Generation
Digital humans are fundamental to immersive interaction, yet creating a unified model for holistic modalities, including text, audio, motion, and visual content, remains an open challenge. In this paper, we present Archon, a fully pretrained, human-centric unified multimodal model for holistic avatar generation. Archon unifies seven modalities with modality-specific tokenizers, and a native autoregressive unified multimodal model pretrained on synchronized modalities and 72 diverse tasks to model holistic joint distributions. To address the token explosion challenge in high-fidelity talking videos, we introduce a memory-efficient semantic video reparameterization, achieving 4x token reduction while preserving fine-grained dynamics, coupled with a semantic-driven video diffusion decoder. We further propose a "Thinking in Modality" that decomposes ambiguous cross-modal tasks into stepwise thinking in an alternative chain of modality, progressively enhancing fidelity and controllability. Extensive experiments demonstrate that Archon achieves superior or comparable performance across diverse digital human generation tasks, validating the effectiveness of our unified framework. Project page: https://zju3dv.github.io/archon/.