AI for Bio · Model

CellFM

Human-only single-cell foundation model with 800 million parameters

Single-cell & perturbationMixedChinaActive中文

Overview

NameCellFM · aliases: CellFM
Full nameThe name is an acronym compound with no expanded form in the paper
Ambiguous names (not auto-linked)None
FamilyNot a family entry. A single model with no family of versions
OrganizationSun Yat-sen University (computer science), Chongqing University, Huawei, Singleron Biotechnologies
Organization typeMixed
CountryChina
Model classSingle-cell foundation model (linear-complexity backbone)
DomainSingle-cell & perturbation
First release · paper revision
Basis: Preprint posted (registry record) source
Other milestones
  • First public evidence: · Preprint posted source
  • Preprint v1: · Preprint v1 source
  • Journal online: · Journal publication source
Latest update† unverified · author correction
Basis: Author correction statement in the repository README
An author correction in the repository README: miscalculated averages in several figure panels plus evaluation bugs affecting two baselines on two datasets, with the authors stating the conclusions are unchanged. The earlier event is the journal publication on 20 May 2025.
Note: No independent permalink for the commit day was obtained, so the date is unverified, and no formal journal erratum was found source
The most recent sourced event, not a claim that nothing changed since.
StatusActive · An author correction statement within 12 months
OpennessCode: Open
Weights: Open
Code in a public repository with a long-term archive and weights in a public model hub; licence text not retrieved this round
Benchmarks reported1 counted (modified protocol 1)
Wet-lab validationNone recorded
Unverified fields† latest_update, open_status.code_license, open_status.weights_license
As of2026-11-16

What it does

Pretrained on human single-cell data only, so that mixing species does not drag down performance on human cells, with a linear-complexity sequence backbone and downstream tasks spanning cell-type annotation, perturbation prediction, gene function prediction and gene relationship capture; training ran on a domestic AI framework and accelerators. It follows the same route as scGPT/Geneformer, distinguished by human-only data and domestic compute.

Benchmarks reported

Every row is one pairing from the survey's relation file; a score belongs to the benchmark page's ranking, where it is shown with metric, eval set, source type and date. 1 of 1 pairings are counted.

BenchmarkRelationEval set / protocolSource typeNote and evidence
24 Virtual Cell Challenge and perturbation benchmarksDeveloper report, modified protocol—Self-reportedAdamson and Norman data through GEARS, not VCC. · evidence

Other evaluations (outside the 44)

  • Cell annotation and gene function prediction (Public benchmark not among the 44) Self-reported: Compared with two single-cell foundation models on public datasets, reporting relative gains in mean accuracy
  • Third-party perturbation evaluations (Public benchmark not among the 44) Third-party: Two independent evaluations find simple baselines competitive with such foundation models; neither includes this model

Developers report: human-only pretraining on about a hundred million cells with a linear-complexity backbone, beating comparable models on annotation, perturbation prediction and gene function prediction, mostly as single-digit relative gains.

Independent evidence: No independent evidence found. Two third-party evaluations do not include this model, and no independent evaluation of it was found

Training data

Group in Figure 7: Public data only.

ClassValueDetail
a. Public databasesYes19,914 samples and about 102.3 million human cells from public databases source
b. Private or licensed dataNo
c. Own wet-lab data used for trainingNo
d. Wet lab used only for validationNo

Wet-lab validation

None. The developers report no wet-lab validation.

Background

  • Organization: An academic team in a Chinese university's computer science school, with collaborators at another university, a laboratory, a telecommunications and compute company, an affiliated hospital and a single-cell sequencing company.
  • Funding: The acknowledgements list national key R&D programme, national natural science foundation, central university research funds, postdoctoral funds and a provincial natural science foundation.
  • Products and partnerships: Code sits under a public repository organisation with a weights card in a public model hub; comparable models include scFoundation and scGPT/Geneformer.
  • Related models: scGPT / Geneformer, xTrimoSCPerturb / scFoundation, STATE

Reading notes

  1. All performance claims are self-reported as relative gains over the runner-up, so the absolute gaps are small.
  2. Independent work finds simple baselines competitive with such foundation models on perturbation prediction, and that doubt applies here too.
  3. The authors posted their own correction of figure and evaluation bugs in the repository, which is commendable, though no formal journal erratum was found.
  4. Its value lies more in human-only data and the feasibility of domestic compute than in a verified predictive edge.

Caveats

  1. No independent permalink was obtained for the correction commit's day.
  2. No formal journal erratum was found; the correction appears only in the repository README.
  3. It has no result on the public blind-test challenge.
  4. Licence text was not retrieved this round.
  5. The perturbation evaluation uses early public datasets, which differ from current blind-test protocols.

Sources

  1. Journal paper 2025 (journal)
  2. Preprint 2024 (preprint)
  3. Code repository and correction statement (repository)
  4. Third-party perturbation evaluation (third-party paper)

Corrections

1 correction
FieldWasNowBasisFound
latest_update2025-05-20 (journal publication)2026-09-07 (author correction notice in the repository); event kind changed to author correctionsource2026-11-16

Part of the AI for Bio benchmarks survey · all 112 models · open in the explorer. Record as of 2026-11-16.