Hugging Face has released a technical report introducing Gander, an end-to-end model designed to unify omni perception, real-time interaction, and agentic capabilities in a single framework. Unlike conventional turn-based systems, Gander continuously processes streaming inputs across 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, and the model can proactively provide intermediate feedback or ask follow-up questions. To support these capabilities natively, Gander employs two key architectural designs. First, it uses a Cerebellum-Brain collaborative framework: the Cerebellum handles real-time interaction and omni conversational abilities, while the Brain manages complex reasoning and higher-level agentic tasks. These components interact continuously through tool calling and an agent orchestration runtime.
Second, the Cerebellum is built on a streaming Thinker-Talker architecture, where user inputs and model outputs are flattened into an ordered token stream at the chunk level, providing a unified representation for low-latency, continuous interaction.
The authors conducted comprehensive evaluations across four dimensions: conversational ability, omni understanding, interactive capability, and agentic intelligence. Internal human evaluations show that Gander maintains the natural and expressive spoken dialogue capabilities of state-of-the-art open-source models while achieving competitive performance in omni interaction. Gander also demonstrates robustness in challenging real-world scenarios, including background noise, multi-party interactions, and backchannel communication.
Gander, along with its models, code, and data, has been released to facilitate further research and development in the community.