INFevo Announces Completion of Over 100 Million Yuan Pre-A Round, Claiming "a New Record for a Single Financing Round by a Chinese Virtual Cell Company"
IT Home, November 16: INFevo, an AI for Science (AI4S) company centered on virtual cells and recursive scientific discovery, announced today the completion of over 100 million yuan in Pre-A funding, setting a new record for the amount raised in a single financing round by a Chinese virtual cell company.

This round was jointly invested by institutions including Xinglian Capital, Prosperity7 Ventures, Huagai Capital, Huakong Fund and DT Capital, while existing shareholders Shunwei Capital, Sequoia China and Green Pine Capital continued to increase their investments.
This is INFevo's second financing round within five months, following its angel round of tens of millions of yuan completed in June 2026. The funds will accelerate the company's AI4S infrastructure development, drive deep coordination among cell response prediction, scientific hypothesis testing and high-throughput automated experimental feedback, and build a full-chain AI biology platform covering "target discovery — virtual screening — clinical protocol optimization," accelerating the translation of frontier computation into precise disease intervention and innovative clinical therapies.
The application of AI in biomedicine is expanding from molecular design to dynamic research at the cellular and systems biology levels, in order to more efficiently explore how interventions such as drugs and genes change complex living systems. However, as the exploration space expands exponentially, a severe mismatch in efficiency and throughput has emerged between the ability of computational systems to generate hypotheses and the capacity of traditional wet-lab validation.
Facing the wet-lab validation bottleneck, AI-driven virtual cell models and high-throughput automated laboratories are expected to become one of the key technologies for breaking through. Models can predict cellular changes caused by drug or gene interventions before experiments, and thereby infer the effects of drugs in more complex biological systems, including animals and humans, helping researchers screen candidate molecules, compare potential targets and set experimental priorities, so that limited experimental resources are used to test more valuable directions, thereby greatly improving the translation efficiency of early discovery and accelerating the process of target discovery and biological mechanism elucidation.

As AI and automation technologies drive upgrades in R&D efficiency, deeper bottlenecks in the research system are also becoming more prominent: the knowledge, analytical methods and negative-result experience generated during scientific discovery are often scattered and lost across isolated projects, making it difficult to accumulate them into reusable organizational assets, which keeps scientific exploration trapped in inefficient repeated trial and error for a long time. How to enable these intermediate research results to break out of silos, achieve scaled accumulation, and be continuously used for model iteration, hypothesis generation, and experimental design and validation has become the key to AI4S moving from point-level efficiency gains to a leap in systematic discovery capability.
In response, INFevo proposes the concept of Recursive scientific discovery (RSD). Recursive self-improvement (RSI) focuses on improving the capabilities of AI systems themselves, while RSD extends continuous improvement to the entire process of scientific discovery.
Through structured retention, association and reuse of scientific questions, hypothesis reasoning, analytical methods and experimental results, this system enables recorded evidence and experience to continuously drive subsequent hypothesis generation, experimental design and candidate screening.

Based on the RSD framework, INFevo has built a dual-driven closed-loop system of "AI foundation large models + high-throughput automated experiments."
- At one end of this system is INFevo's self-developed foundation model OCOO, covering cellular multi-omics, used to predict cellular responses caused by interventions such as drugs and genes;
- At the other end is an automated wet-lab system covering cell culture, perturbation and multi-omics detection, used to test key predictions and generate new data.
The model helps determine more promising experimental directions, while experimental results are used to test predictions, improve the model and guide subsequent research. The two are connected through a unified data and research artifact management mechanism, allowing the evidence and experience accumulated in each exploration to continuously improve the system's prediction and experimental design capabilities.
Relying on the above underlying architecture, within months of its founding INFevo has achieved leading results in core scenarios such as target discovery, virtual screening and preclinical evaluation, driving the relevant capabilities from model R&D into real research scenarios.
In terms of model R&D, the company has released OCOO-T, a virtual cell foundation model in the OCOO series based on transcriptomic data, which directly models changes in gene expression of cells under various perturbations. With a concise and scalable model architecture, it achieves unified modeling across perturbation types and leading predictive performance. In same-condition evaluations on three authoritative single-cell perturbation benchmarks — Tahoe100M, Replogle and PBMC — OCOO-T achieved industry-leading performance, laying the foundation for its use in simulating real perturbation biology and supporting target discovery and early efficacy evaluation.