FrontierAgent
Open-source terminal agent runtime: a stateful ReAct agent or a coordinator with parallel sub-agents, working in a sandboxed filesystem for long-horizon research and file-based deliverables.
Relax Omni-Modal RL Engine
An asynchronous distributed reinforcement-learning engine for post-training text, vision, and audio models.
LLM-in-Sandbox
Put your LLM inside a Docker-isolated code sandbox and it beats the standalone model on math, physics, chemistry, long-context and instruction-following tasks — no extra training required, and RL amplifies the gains further.
Functionary: Conversational LLM for Controlled Tool Use
Functionary is a language model that can interpret and execute functions/plugins, deciding when to call tools and understanding their results.
AReaL Async RL Platform
Asynchronous RL infrastructure for training reasoning models and tool-using agent workflows.