Gander unifies real time multimodal interaction and agent reasoning
Omni Interaction Agent Technical Report
Artificial IntelligenceMachine LearningMultimediaSound
Summary
Interacting naturally with AI often means waiting for one response at a time, which can feel slow and awkward. The authors created Gander, a system that listens to and understands video, speech, and text all at once, allowing it to have smooth back-and-forth conversations and handle complex tasks. It can respond instantly or ask questions anytime during an interaction. The system splits duties: one part manages real-time chatting and the other handles deeper thinking and planning. Tests show Gander works well even in noisy places or with multiple people talking.
What this means in practice
- •For customer support teams: Enable AI agents to handle live, multimodal customer conversations without turn-taking delays, improving response speed and naturalness.$Commercial implications: This approach allows companies to offer more fluid and responsive AI chat assistants across speech, video, and text channels.
- •For remote collaboration software developers: Integrate continuous multimodal AI interaction that supports interruptions and proactive feedback during complex remote workflows.
Authors
Orantqing, Shengpeng Ji, Junlong Tong, Jialong Zuo, Dongjie Fu, Di Cao, Yangzhuo Li, Shangda Wu, Franz, Evan, Theron Veyra, Changhao Pan, Jingyu Lu, Dongchao Yang, Zhifei Xie, Yang Tan, Xiaoyu Shen, Xiaoda Yang, Wenfu Wang, Teddysun, Steveyves, Zhou Zhao, Bryanytian
Abstract
In this work, we present Gander, an end-to-end model that unifies omni perception, realtime interaction, and agentic capabilities within a single framework. In contrast to turn-based conventional paradigms, Gander continuously receives streaming inputs across multiple modalities, including video, speech, and text, enabling natural full-duplex interaction in both everyday conversations and complex workflow-oriented agent scenarios. Users can interrupt the model at any time, while the model can also proactively provide intermediate feedback or ask follow up questions. To natively support these capabilities, Gander adopts two key architectural designs: 1) It employs a Cerebellum-Brain collaborative framework, in which the Cerebellum is responsible for realtime interaction and omni conversational capabilities, while the Brain handles complex reasoning and higher-level agentic tasks. The two components interact continuously through tool calling and the agent orchestration runtime. 2) The Cerebellum is built upon a streaming Thinker-Talker architecture, user inputs and model outputs are further flattened into an ordered token stream at the chunk level, providing a unified representation for low latency, continuous interaction. We conduct comprehensive evaluations of Gander across four dimensions: conversational ability, omni understanding, interactive capability, and agentic intelligence. Internal human evaluations demonstrate that Gander maintains the natural and expressive spoken dialogue capabilities of SOTA open source models while achieving competitive performance in omni interaction. Gander also demonstrates robustness in challenging real-world scenarios, including background noise interference, multi-party interactions, and backchannel communication. We release Gander together with its models, code, and data to facilitate further research and development in the community.