← Single-cell & foundation models
Stack
Arc Institute model trained on 149M human cells using tabular attention for in-context learning; used to generate the Perturb Sapiens atlas (28 tissues, 201 perturbations).
| Category | Single-cell & foundation models |
|---|---|
| GitHub | ArcInstitute/stack |
| First public | 2026-01-09 (source) |
| Paper | https://doi.org/10.64898/2026.01.09.698608 |
| License | CC-BY-NC-SA-4.0 (code); Arc non-commercial (model) |
| Pricing | Open source (non-commercial) |
| Tags | #foundation-model#in-context-learning#perturbation |
| Last checked | 2026-10-16 |
Specs
| Pretraining corpus | 149M human cells |
|---|---|
| Architecture | Tabular-attention transformer |
| Tasks | Zero-shot embedding, in-context perturbation prediction |
| Species | — |
| GitHub stars | — |
| Interface | — |
Specs come from each project's README, paper or official page · Stack paper · Stack README · Stack source 3. “—” means we couldn't verify it.
Compare with…
Pick 1–3 other Single-cell & foundation models tools to compare side by side with Stack.
Related
Generative pretrained transformer foundation model for single-cell multi-omics.
Single-cell & foundation modelsOpen source#foundation-model#transformer
Transformer pretrained on ~30M single-cell transcriptomes for network biology predictions.
Single-cell & foundation modelsOpen source#foundation-model#transfer-learning
Probabilistic deep generative models for single-cell omics (scVI, totalVI, scANVI...).
Single-cell & foundation modelsOpen source#vae#integration