HybridAgent: Think in Latent, Act in Text

How can we build an agent that thinks beyond human language, yet still acts through it in the external environment? Modern LLM agents still reason and interact through text, while latent reasoning offers an efficient internal medium but lacks native tool access and reasoning transparency. We bridge these two directions with HybridAgent, which dynamically navigates between latent reasoning and text-based interaction throughout a unified generation process. Architecturally, a lightweight HybridLink enables seamless transitions between text and latent modes, while the agent's continuous working memory preserves lossless hybrid context across multiple turns. These designs preserve tool-integrated agentic capabilities while making latent thoughts directly inspectable. Building on this architecture, we further develop dedicated post-training algorithms for hybrid reasoning: HybridSFT first stabilizes tool-integrated hybrid generation by teaching the backbone to coordinate with HybridLink, while HybridRL further scales reasoning by learning when to think in latent space and when to act through text. Empirically, we evaluate HybridAgent at 3 model scales across 11 benchmarks covering mathematics, science, medicine, search, and agentic reasoning. Compared with frontier LLM Agents and agentic frameworks spanning both text-based and latent methods, HybridAgent consistently improves performance by 10.4%, while delivering 35.5%–54.7% reduced token overhead, 1.8x–2.6x faster inference, and 2.7x–3.7x cost savings.