BioSimBench
BioSimBench is a datasets & benchmarks tool, first public in 2026. CellType benchmark that scores language models on predicting held-out cellular responses to drugs, gene knockdowns and immune signals against real lab measurements; paired with BioSimGym RL training environments.
| Category | Datasets & benchmarks |
|---|---|
| Website | www.celltype.com/biosimbench |
| Released | 2026-11-04 (source) |
| License | public leaderboard and evaluation site; no SPDX license file confirmed from a primary source as of 2026-11-05 |
| What this allows | Unknown No confirmed terms from a primary source; ask the authors before using it in a product. What do these mean? |
| License source | www.celltype.com/biosimbench · checked 2026-11-05 |
| Pricing | No cost (research use) |
| Tags | #benchmark#perturbation#virtual-cell#drug |
| Last checked | · links verified |
Can I use BioSimBench commercially?
Unclear. We could not confirm the terms from a primary source; check with the authors before commercial use.
- License: public leaderboard and evaluation site; no SPDX license file confirmed from a primary source as of 2026-11-05
- Source: www.celltype.com/biosimbench · license checked 2026-11-05
What the license types mean · not legal advice.
Specs
| Size | About 60 cases per prediction track (immune 27; drug-response extended panel 820); 296 research-agent episodes |
|---|---|
| Modality | — |
| Access | Web leaderboard at celltype.com/biosimbench |
| Interface | — |
Compare with…
Pick 1–3 other Datasets & benchmarks tools to compare side by side with BioSimBench.
Related
Meta FAIR's open dataset of DFT calculations on diverse molecules, with evaluations and baseline models.
Datasets & benchmarksOpen sourceUnknown#dft
ML-ready datasets and benchmarks for therapeutics.
Datasets & benchmarksOpen sourcePermissive (commercial OK)#benchmark
Benchmarks for protein fitness and mutation effect prediction.
Datasets & benchmarksOpen sourcePermissive (commercial OK)#benchmark