Corollary

Research

  • Ask

Library

  • Catalog

Account

  • Overview
  • Jobs
  • Usage
  • Billing
Settings
Corollary
  1. Catalog

Library

Find the model or dataset for the question you have.

Ask in plain language and we will work out where the answer lives — or filter the index yourself. Almost everything here runs on our servers, from this page, with no setup.

Models
297
Datasets
245
Organisations
128

Labs, institutes and companies

Ready to run
24

Including 6 that run free and instantly

Looking for a particular value, not a particular dataset?Search schemas, rows and column statistics inside the data itself.

Things that work

Filtering

212 entries in Chemistry

Showing 48 of 212

Ready to run

9
InstantModel

Molecular Property Calculator

Corollary Labs

Lipinski, Veber and lead-likeness descriptors computed from SMILES — instantly, free.

property calculation
ReadyModel

BioMedGPT-R1

PharMolix / OpenBioMed

Reasoning-capable biomedical LLM with chain-of-thought over molecules.

chat
ReadyModel

ChemBERTa-2

Seyone Chithrananda et al.

RoBERTa over SMILES — the fast, widely validated chemistry baseline.

property prediction
ReadyModel

DiffDock

MIT / Barzilay Lab

Diffusion docking: 38% top-1 success versus 23% for classical AutoDock.

molecular docking
ReadyModel

Galactica

Meta AI

Scientific LLM retained as a domain-knowledge evaluation baseline only.

evaluation baseline
ReadyModel

MAMMAL

IBM Research

One model across proteins, molecules and omics — SOTA on 9 of 11 discovery tasks.

multi task
ReadyModel

MolFormer

IBM Research

SMILES transformer pre-trained on 1.1 billion molecules.

property prediction
ReadyModel

MolMIM

NVIDIA BioNeMo

Controlled molecule generation in a well-behaved latent space.

molecular generation
ReadyModel

TxGemma-27B

Google Health AI

Open therapeutics LLM covering 66 Therapeutic Data Commons tasks.

chat

Everything else indexed

39
ABMIL (CONCH v1.5 features, pc108-24k)MahmoodLabAttention-based MIL classifier on CONCH v1.5 features — pan-cancer pc108-24k training for slide-level subtyping and downstream pathology evaluation.image-feature-extractionOn demand
ABMIL (UNI features, pc108-24k)MahmoodLabAttention-based MIL classifier on UNI features — pan-cancer pc108-24k training; provides slide-level aggregation over the original UNI patch encoder.image-feature-extractionOn demand
ABMIL (UNI2 features, pc108-24k)MahmoodLabAttention-based multiple-instance learning classifier on UNI2 features — pan-cancer pc108-24k training; pairs with the UNI2-h patch encoder for slide-level inference.image-feature-extractionOn demand
ADMET-AISwanson et al., Stanford52 ADMET endpoints in one call — hERG, BBB, CYP, clearance, solubility, Tox21.admetOn demand
AIFS Single 1.0ecmwfFirst 1.0 release of the deterministic AIFS — graph-based global weather forecasting model from ECMWF on the Anemoi framework.forecastingOn demand
AIMNet2 (B97-3c, 2025)isayevlabAIMNet2 retrained at the B97-3c level of theory — 2025 release with improved coverage and accuracy.property predictionOn demand
AIMNet2 ωB97M-D3isayevlabNeural network interatomic potential for fast and accurate molecular simulations, trained at the ωB97M-D3 level of theory.property predictionOn demand
AIMNet2-NSEisayevlabAIMNet2 specialised for open-shell chemistry (radicals, transition states) — neural network interatomic potential for non-singlet electronic states.property predictionOn demand
AIMNet2-PdisayevlabAIMNet2 specialised for palladium-containing organometallic systems — supports homogeneous catalysis simulation at near-DFT accuracy.property predictionOn demand
AIMNet2-rxnisayevlabAIMNet2 trained on reaction data — neural-network interatomic potential supporting reactive molecular simulations.property predictionOn demand
AutoDock VinaScripps ResearchThe docking engine every other docking paper compares itself against.molecular dockingOn demand
BioCLIP 2imageomicsOpenCLIP-based foundation model for organismal biology — zero-shot species classification from photographs across the tree of life, trained on TreeOfLife-200M.zero-shot-image-classificationOn demand
Chemprop D-MPNN — toxicity panelSwanson et al., Stanford · Chemprop, MITThe toxicity half of the ADMET panel: Ames, hERG, DILI, ClinTox, LD50 and all twelve Tox21 assays.toxicityOn demand
CONCHMahmoodLabContrastive vision-language model for pathology trained on 1.17M histology image–caption pairs — enables zero-shot classification, retrieval, and report-grounded analysis on H&E whole-slide images.image-feature-extractionOn demand
CONCH v1.5MahmoodLabUpdated CONCH checkpoint (v1.5) with strengthened image-side encoder and expanded pretraining corpus — drop-in upgrade for zero-shot and few-shot pathology tasks.image-feature-extractionOn demand
DoctoBERT-fr-basedoctolib-labFrench medical encoder (RoBERTa, 111M params, 512-token context) pretrained from scratch on FineMed-fr — state-of-the-art on DrBenchmark, leading five of seven tasks across French and English medical encoder families.fill-maskOn demand
DoctoModernBERT-fr-basedoctolib-labFrench medical encoder (ModernBERT, 149M params, up to 8,192-token context) pretrained from scratch on FineMed-fr — best on real-world clinical NER and top-ranked on DrBenchmark. ## Blog Posts (18)fill-maskOn demand
ESM3EvolutionaryScaleSequence, structure and function unified in one generative transformer.generationOn demand
ESMFold2biohubState-of-the-art protein structure prediction and design model — defines a new frontier for speed and accuracy, with unprecedented success rates for protein binder and antibody generation.feature-extractionOn demand
ESMFold2-ExperimentalbiohubExperimental ESMFold2 variant exploring new training recipes and architectural changes for protein structure prediction, design, and confidence estimation.feature-extractionOn demand
ESMFold2-Experimental-FastbiohubFast variant of the experimental ESMFold2 line — combines accelerated inference with the research-recipe improvements of the experimental branch.feature-extractionOn demand
ESMFold2-FastbiohubAccelerated variant of ESMFold2 — optimised for high-throughput structure prediction across large protein sets while retaining strong accuracy.feature-extractionOn demand
eva-rnaScientaLabTransformer foundation model producing sample-level and gene-level embeddings from RNA-seq profiles (bulk, microarray, pseudobulked single-cell) in human and mouse.feature-extractionOn demand
FlashPPItattabioFast protein-protein interaction prediction model — trained for high-throughput screening of interaction networks.feature-extractionOn demand
GENA-LM BERT base (T2T)AIRI-InstituteBERT-base-style genomic foundation model trained on T2T assemblies — lighter-weight backbone for genomic sequence understanding.fill-maskOn demand
GENA-LM BERT large (T2T)AIRI-InstituteBERT-large-style genomic foundation model trained on telomere-to-telomere human assemblies — supports variant interpretation, regulatory prediction, and downstream genomic tasks.fill-maskOn demand
gLM2 650Mtattabio650M-parameter genomic foundation model from Tatta Bio — trained on the OMG open-mixed-genomes corpus for sequence-level biological reasoning.fill-maskOn demand
Indus SDE v0.2nasa-impactScience domain extraction model for identifying and classifying scientific concepts, variables, and entities from geoscience and atmospheric science text.fill-maskOn demand
IsoformerInstaDeepAITransformer model integrating DNA sequence, RNA expression, and protein context for isoform-level gene expression prediction.fill-maskOn demand
MACE-MH-0mace-foundationsMACE foundation model targeting molecular and hybrid systems — equivariant potential trained on a unified molecular/materials dataset.property predictionOn demand
MACE-MH-1mace-foundationsUpdated MACE-MH foundation potential with refined molecular/materials hybrid training — successor to MACE-MH-0.property predictionOn demand
MACE-MP-0mace-foundationsMACE foundation model trained on the Materials Project — equivariant message-passing potential for inorganic crystal simulation across most of the periodic table.property predictionOn demand
MACE-MPA-0mace-foundationsMACE foundation model trained on the Materials Project + Alexandria datasets — broader coverage variant for inorganic-materials simulation.property predictionOn demand
MedASRgoogleMedical automatic speech recognition model for clinical documentation.automatic-speech-recognitionOn demand
MedSigLIPgoogleMedical image-language model for visual understanding in healthcare.zero-shot-image-classificationOn demand
MIST 1.8B · ⟨R²⟩mist-modelsMIST 1.8B fine-tuned for electronic spatial extent from QM9.feature-extractionOn demand
MIST 1.8B · BACEmist-modelsMIST 1.8B fine-tuned on BACE — Alzheimer-target inhibition classification.feature-extractionOn demand
MIST 1.8B · BBBPmist-modelsMIST 1.8B fine-tuned on BBBP — blood-brain-barrier permeability.feature-extractionOn demand
MIST 1.8B · ClinToxmist-modelsMIST 1.8B fine-tuned on ClinTox — clinical toxicity classification.feature-extractionOn demand

164 more behind this view