OmniGAIA: Towards Native Omni-Modal AI Agents

TL;DR

OmniGAIA introduces a new benchmark and foundation model for omni-modal AI agents capable of deep reasoning and multi-turn tool use across vision, audio, and language modalities, advancing towards general AI assistants.

Contribution

The paper presents OmniGAIA, a comprehensive omni-modal benchmark, and OmniAtlas, a native omni-modal foundation agent with active perception and tool integration capabilities.

Findings

OmniAtlas improves multi-modal reasoning accuracy.

OmniGAIA enables evaluation of complex cross-modal tasks.

OmniAtlas enhances tool-use capabilities of open-source models.

Abstract

Human intelligence naturally intertwines omni-modal perception -- spanning vision, audio, and language -- with complex reasoning and tool usage to interact with the world. However, current multi-modal LLMs are primarily confined to bi-modal interactions (e.g., vision-language), lacking the unified cognitive capabilities required for general AI assistants. To bridge this gap, we introduce OmniGAIA, a comprehensive benchmark designed to evaluate omni-modal agents on tasks necessitating deep reasoning and multi-turn tool execution across video, audio, and image modalities. Constructed via a novel omni-modal event graph approach, OmniGAIA synthesizes complex, multi-hop queries derived from real-world data that require cross-modal reasoning and external tool integration. Furthermore, we propose OmniAtlas, a native omni-modal foundation agent under tool-integrated reasoning paradigm with active omni-modal perception. Trained on trajectories synthesized via a hindsight-guided tree exploration strategy and OmniDPO for fine-grained error correction, OmniAtlas effectively enhances the tool-use capabilities of existing open-source models. This work marks a step towards next-generation native omni-modal AI assistants for real-world scenarios.