Launching today
OpenScience
The open-source AI workbench for scientific research
11 followers
The open-source AI workbench for scientific research
11 followers
OpenScience is an open-source AI agent for scientific research. It reads papers, writes and runs code, works in notebooks and a terminal, and sends long jobs to Modal, your own servers, or a Slurm/PBS cluster. Give it a metric and Autoresearch runs experiments, logs every one, and keeps hill-climbing. Use 30+ models through one wallet, sign in with ChatGPT, or run a local model. It ranks #1 on agentic scientific research benchmarks. Free and open source.
Hey Product Hunt, I'm Ishaan, one of the people building OpenScience.
We started this because doing research with AI meant copying things back and forth between a chat window, a terminal, a notebook, and whatever cluster you had access to. OpenScience puts all of that in one place and lets an agent work across it.
We also think scientific tools should be open. However, the strongest models and products increasingly come with restrictions set by the labs that train them: which fields they'll help with, who gets access, and on what terms. So OpenScience is open source and model-agnostic. Pick any model, swap it mid-project, or run it locally on your own hardware.
OpenScience is currently #1 on agentic scientific research benchmarks, including Terminal Bench Science, TerminalBench 4.0 (Science subset), BiomniBench, and our internal OpenScience Bench.
What's in this launch:
1. New IDE: a faster, redesigned workspace with papers, code, notebooks, a LaTeX editor, a terminal, and 3D protein and genomics viewers next to the agent
2. Autoresearch: experiment loops that run on Modal GPUs, your own servers, or Slurm and PBS clusters
3. OpenScience Ace: 30+ models (GPT-6 Astra, Claude Opus 5.5, Kimi K3, GLM-5.3 and more) through one pay-as-you-go wallet, or bring your own keys, your ChatGPT/Codex login, or a local model
4. Updated tooling: 300+ research skills, 50+ scientific tools and databases, and NVIDIA BioNeMo support
Not a scientist? Autoresearch is probably what you'll care about. Give it a metric, like validation loss, p99 latency, or build time, and it forms hypotheses, runs experiments, and logs every attempt so you can see what worked.
OpenScience is built for computational research across ML, biology, chemistry, physics, and data science. Researchers at 30+ universities and labs used the beta, and now it's open to everyone. Sign up with a .edu email for $5 in free Ace credit.
I'll be in the comments all day, would love to hear what you think!