#6161st of 92 in Agent platformsFor science, health and research, science, research and papers
openscience
Research agent that reads papers, runs code and experiments, and logs every step
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.4k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- TypeScript
Missed the latest GitHub check; these facts may be older.
Overview
OpenScience is a research agent with a workbench around it, shipped as an Electron desktop app, a browser workspace and a CLI. It plans a task, searches literature and databases such as PubMed, arXiv, ChEMBL and UniProt, runs code in shell, Python and R kernels, and returns results with a full trace. Compute can be local, Modal, SSH hosts, or Slurm and PBS clusters, with approval before jobs run.
Who it is for: Scientists and ML researchers doing literature review and data analysis
Strengths
- Full trace of searches, code runs and file writes for each turn
- Hundreds of bundled science skills plus connectors to PubMed, arXiv, ChEMBL, UniProt
- Remote compute on Modal, SSH, Slurm or PBS, approved before jobs run
- Works with own API keys, local models via Ollama or LM Studio, or Ace
Weaknesses
- First-run setup asks you to sign in to an OpenScience account
- Skills are mostly third-party; a skill does not guarantee its tools are installed
- NVIDIA BioNeMo adapters are experimental and need your own NVIDIA key
- Published benchmarks use GPT-6 models, not open-weight ones
What it needs
- GPU optional
- Needs Node/npm (CLI install), Bun (build from source), Modal, SSH or Slurm/PBS (optional remote compute)
- Models: Provider API keys, Ollama, LM Studio, OpenAI-compatible local endpoints, OpenRouter
Also in Agent platforms
See all 92| 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 |