---
title: rbio1
description: rbio1 is a 3B biology reasoning model from the Chan Zuckerberg Initiative, post-trained from Qwen2.5-3B-Instruct by reinforcement learning (GRPO) with virtual-cell models and the Gene Ontology as soft verifiers. It comes in about ten variants, one per verifier. The weights are distributed by CZI's Virtual Cells Platform, not on Hugging Face; this repository holds the code.
tags:
- reasoning
- single-cell
- source-code
- virtual-cell
license: other
source_url: https://github.com/czi-ai/rbio
---

# rbio1

> **Mirrored metadata.** This card is reproduced on Atlas for discovery. The artifact itself is not hosted here — follow the source link for the files, and refer to the upstream licence for terms of use.

rbio1 is a 3B biology reasoning model from the Chan Zuckerberg Initiative, post-trained from Qwen2.5-3B-Instruct by reinforcement learning (GRPO) with virtual-cell models and the Gene Ontology as soft verifiers. It comes in about ten variants, one per verifier. The weights are distributed by CZI's Virtual Cells Platform, not on Hugging Face; this repository holds the code.

Upstream repository: `czi-ai/rbio`. Weights: [CZI Virtual Cells Platform](https://virtualcellmodels.cziscience.com/model/rbio). Licence: CZI's own terms, derived from the Qwen licence.

**Source:** https://github.com/czi-ai/rbio

**Licence:** other

_This model is part of the Atlas demo catalogue. See the repository history for provenance._