▶AI Research Directory
in

InfiAgent

Agent framework for long-running tasks

Open-sourceResearch Agents & AI ScientistsLLM as Scientist

AT A GLANCE

Intended use: Building configurable agents for research and other extended workflows.

Pricing model: Open-source. Open-source availability does not establish free hosted usage, model access, or compute.

Public-source review: . Documentation checked; product performance was not tested.

LIMITATIONS

  • Long-horizon execution does not remove the underlying model's context limits.
  • Requires runtime and model-provider setup; review tool access and running costs.

FEATURES

  • ▸ Supports hierarchical or flat agent configurations and reusable skills.
  • ▸ Persists work in files and supports resuming interrupted tasks.

What it is

An agent framework for long-horizon tasks, using configurable agent hierarchies, file-based memory, skills, and resumable execution rather than an unlimited model context window.

RELATED · SAME CATEGORY