Back to directory

OpenDrSai

A Chinese research institute's framework for building supervised, reproducible scientific AI agents and multi-agent systems, built on Microsoft AutoGen and organized around a brain, actuators, memory, and sensors architecture.

OpenDrSai

OpenDrSai is a development and deployment framework for scientific agents and multi-agent systems, built by the HEPAI team at the Institute of High Energy Physics. It organizes agent capability around a BAMS model: a brain for reasoning and decisions, actuators for tool use and task execution, memory for long and short-term context and state, and sensors for user input, scientific data, and external services and models.

The framework runs on Microsoft AutoGen as its core agent runtime and integrates with the HepAI platform for model access, identity, and worker deployment, with optional connections to RAGFlow or ChromaDB for knowledge-base and long-term memory. Agents can be deployed as OpenAI Chat Completions-compatible endpoints or HepAI workers, and are used through a web UI, terminal interface, desktop client, or mobile app, aimed at multi-task, long-running, human-in-the-loop scientific work.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.