#2727th of 80 in Agent platformsFor science, health and research, research and papers
openresearch
Desktop app and CLI that runs research agents with experiment tracking
- Stars
- 8.1k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Rust
9 on 2 Jun 20268,070 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OpenResearch is a desktop app and `orx` CLI that wraps coding agents (Claude Code, Codex, OpenCode, Cursor, Google Antigravity) in a research workflow. Agents search alphaXiv, bioRxiv and PubMed, then run experiments locally or on SSH, Slurm, Kubernetes, Modal, Tinker, Ray or Hugging Face Jobs. Every experiment, with its logs, code and artifacts, is stored in a local SQL database that agents can consult later.
Who it is for: ML researchers running agent-driven experiments on local or cluster compute
Strengths
- Works with several coding agents, plus local models via OpenCode with LM Studio, oMLX or Ollama
- Many compute backends: local, SSH, Slurm, Kubernetes, Modal, Ray, Hugging Face Jobs
- Experiment logs, code and artifacts stay in a local SQL database on your machine
- Literature search built in for alphaXiv, bioRxiv and PubMed
Weaknesses
- Official builds send opt-out analytics and agent feedback; disable in Settings or `orx telemetry off`
- Windows build is beta and requires Git for Windows
- CLI installed by install.sh is unsigned and may be blocked on managed Macs
- Managed compute and email updates need an openresearch.sh account
What it needs
- Needs Git for Windows (Windows), a coding agent such as Claude Code, Codex or OpenCode
- Models: Claude Code, Codex, OpenCode, Cursor, Google Antigravity
- port 4791
Quick answers
Is openresearch open source?
Yes. Its code is on GitHub under the MIT license.
Also in Agent platforms
See all 80| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |