Gradientsys: A Multi-Agent LLM Scheduler with ReAct Orchestration
This addresses the challenge of efficient and transparent multi-agent orchestration for AI applications, though it appears incremental as it builds on existing scheduling and planning concepts.
The paper tackles the problem of coordinating diverse AI agents for complex tasks by introducing Gradientsys, a multi-agent scheduling framework that uses an LLM-powered scheduler and ReAct-based planning, achieving higher task success rates with reduced latency and lower API costs on the GAIA benchmark compared to a baseline.
We present Gradientsys, a next-generation multi-agent scheduling framework that coordinates diverse specialized AI agents using a typed Model-Context Protocol (MCP) and a ReAct-based dynamic planning loop. At its core, Gradientsys employs an LLM-powered scheduler for intelligent one-to-many task dispatch, enabling parallel execution of heterogeneous agents such as PDF parsers, web search modules, GUI controllers, and web builders. The framework supports hybrid synchronous/asynchronous execution, respects agent capacity constraints, and incorporates a robust retry-and-replan mechanism to handle failures gracefully. To promote transparency and trust, Gradientsys includes an observability layer streaming real-time agent activity and intermediate reasoning via Server-Sent Events (SSE). We offer an architectural overview and evaluate Gradientsys against existing frameworks in terms of extensibility, scheduling topology, tool reusability, parallelism, and observability. Experiments on the GAIA general-assistant benchmark show that Gradientsys achieves higher task success rates with reduced latency and lower API costs compared to a MinionS-style baseline, demonstrating the strength of its LLM-driven multi-agent orchestration.