TL;DR: Apodex released Apodex 1.1, a new model family designed to enhance agentic intelligence for complex, real-world tasks, alongside an open-source agent harness.
Summary: Apodex has launched its Apodex 1.1 model family, specifically engineered to scale agentic intelligence. These models are optimized for sustained, verifiable progress on complex objectives, including reasoning, search, file operations, code execution, failure recovery, and multi-agent coordination. The release includes various model sizes, such as Apodex-1.1-mini, and is accompanied by an open-source agent harness and two research papers: the Apodex 1.1 model paper and the FrontierChallenge benchmark paper.
Why it matters: AI builders can leverage Apodex 1.1 to develop more robust and capable AI agents for intricate workflows. Experiment with the open-source agent harness and models to build applications requiring advanced reasoning and task execution.
Source: reddit