Overview
| Name | C2S-Scale · aliases: C2S-Scale, Cell2Sentence |
|---|---|
| Full name | Cell2Sentence-Scale |
| Ambiguous names (not auto-linked) | None |
| Family | Part of the Cell2Sentence family |
| Organization | Yale University, Google Research, Google DeepMind |
| Organization type | Academic lab |
| Country | United States |
| Model class | Single-cell language model |
| Domain | Single-cell & perturbation (also: Text, LLMs & agents) |
| First release | · preprint Basis: Preprint v1 Note: The site recorded 2025-04, month level only source |
| Other milestones | |
| Latest update | · paper revision Basis: Preprint v4 The preprint was revised to v4 (v2 on 2025-10-11, v3 on 2025-10-26); the earlier event is the 2025-10 release of the largest size. No journal version was found source The most recent sourced event, not a claim that nothing changed since. |
| Status | Active · Preprint revisions and a weights release within 12 months |
| Openness | Code: Open Weights: Open · CC-BY-4.0 Code is in a public repository and weights on the model hub with a Creative Commons licence on the model card; the code licence was not re-checked this round |
| Benchmarks reported | 1 counted (modified protocol 1) |
| Wet-lab validation | 1 recorded campaign |
| Unverified fields | † release_milestones.weights_release, wetlab_validation[0], open_status.code_license |
| As of | 2026-11-16 |
What it does
Ranks each cell's most expressed genes into a text sequence and uses a general language model for cell-type annotation, conditional generation and question answering, extending to spatial neighbourhoods and perturbation response; it keeps the general LLM architecture rather than a single-cell-specific one, the opposite of TranscriptFormer. The developers report that some context-dependent reasoning appears only at the largest size.
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.
| Benchmark | Relation | Eval set / protocol | Source type | Note and evidence |
|---|---|---|---|---|
| 24 Virtual Cell Challenge and perturbation benchmarks | Developer report, modified protocol | — | Self-reported | Uses its own datasets and metrics, not the VCC protocol. · evidence |
Other evaluations (outside the 44)
- Cell-type annotation comparison (Comparison against baselines only) Self-reported: Self-reported parity with two single-cell foundation models on annotation; a comparison rather than a public-protocol evaluation
- Self-chosen metric for integration (Own task set) Self-reported: The preprint does not use the common integration metrics and substitutes another alignment measure, which cannot be listed beside the site's integration evaluation
Developers report: Developers report that handling single-cell data with a general language model improves with scale and can produce biological hypotheses testable in the lab.
Independent evidence: No independent evidence found. The preprint is not peer reviewed and no third-party evaluation of its perturbation or annotation results was found
Training data
Group in Figure 7: Public data + wet lab for validation only.
| Class | Value | Detail |
|---|---|---|
| a. Public databases | Yes | Over 50 million human and mouse cells from public cell atlases and other sources plus biological text and metadata, about 1 billion tokens and 150 million training samples source |
| b. Private or licensed data | No | source |
| c. Own wet-lab data used for training | No | source |
| d. Wet lab used only for validation | Yes | One drug hypothesis was tested in the lab, as validation rather than training source |
Wet-lab validation
Every rate carries its numerator and denominator. Claims that could not be verified are shown as unverified, without numbers; different paper versions are separate entries.
Denominator basis: One hypothesis taken forward from a computational screen of over 4,000 drugs; batch and replicate counts are not public, so no hit rate can be computed
The model predicted that a kinase inhibitor would raise antigen presentation only under low-interferon conditions, and an increase was seen in a previously unseen human cell model; the result exists only in the 2025-10 and later preprint versions, so citations must point at the current version
Executor: Run by the collaborating academic lab; not an independent reproduction
Background
- Organization: A collaboration between a US university lab and two research units of a large technology company, built on that company's open model family.
- Funding: The corporate partner supplied models and compute; no grant numbers are listed on the paper page or the blog.
- Products and partnerships: The model family was released in 2025-04 and the largest size in 2025-10 with an official blog post; the same company's genomics work is AlphaGenome.
- Related models: TranscriptFormer, AlphaGenome, scGPT / Geneformer
Reading notes
- Comparisons are self-reported, with the main text giving directional conclusions and per-item values left to the supplement.
- The drug-hypothesis test is a single case, not systematic wet-lab backing.
- It reports no public integration protocol, so it cannot be compared directly with single-cell models that do.
- The wet-lab result appears only from the 2025-10 version onward, so citing the earlier version misses it
Caveats
- The site linked the first preprint version, which does not contain the wet-lab result; the link must point at the current version.
- Replicate counts and executor attribution are unverified; no replicate figure is given.
- The preprint is not peer reviewed, so all comparison values are the authors' own.
Sources
- Preprint full text (preprint)
- Preprint version record (registry)
- Model hub weights page (model hub)
- Official blog (wet-lab result) (blog)
- Code repository (repository)
Corrections
3 corrections
| Field | Was | Now | Basis | Found |
|---|---|---|---|---|
| first_release | 2025-04 | 2025-04-17 (preprint v1 posted) | source | 2026-11-16 |
| latest_update | 2025-10-15 (weights release, recorded as new weights) | 2026-01-01 (preprint v4, kind paper revision); the weights release is recorded as an earlier event | source | 2026-11-16 |
| wetlab_validation | Cited the first preprint version | Cites the 2025-10 and later versions; the first version does not contain this result | source | 2026-11-16 |
Part of the AI for Bio benchmarks survey · all 112 models · open in the explorer. Record as of 2026-11-16.