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

542 entries

Showing 48 of 542

Ready to run

24
InstantModel

AlphaFold DB Lookup

DeepMind / EMBL-EBI

Fetch a pre-computed structure for any UniProt accession — 247M entries, instantly.

structure retrieval
InstantModel

ESMFold2

EvolutionaryScale / CZ Biohub

Single-sequence folding at state-of-the-art accuracy — no MSA, no waiting.

structure prediction
InstantModel

Molecular Property Calculator

Corollary Labs

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

property calculation
InstantModel

PaperQA2 (Crow)

FutureHouse

Grounded literature retrieval over the biomedical corpus — every claim cited.

literature search
InstantModel

PDB Structure Fetch

RCSB Protein Data Bank

Experimentally determined coordinates and metadata for any PDB entry.

structure retrieval
InstantModel

Sequence Analytics

Corollary Labs

Composition, molecular weight, isoelectric point and hydropathy — deterministic.

annotation
ReadyModel

AlphaFold2

DeepMind (served as BioNeMo NIM)

The reference MSA-based predictor, with managed MSA search included.

structure prediction
ReadyModel

BioMedGPT-R1

PharMolix / OpenBioMed

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

chat
ReadyModel

Boltz-2

MIT × Recursion

Structure and binding affinity in one pass, approaching FEP accuracy 1000× faster.

structure prediction
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

DNABERT-2

Zhou et al., Northwestern

BPE-tokenised genomic BERT for motif, splice site and regulatory classification.

motif discovery
ReadyModel

ESM2-3B

Meta FAIR

High-quality 1,280-dimensional protein embeddings for everything downstream.

sequence embedding
ReadyModel

ESM2-650M

Meta FAIR

The throughput workhorse of the ESM2 family — good embeddings, tiny footprint.

sequence embedding
ReadyModel

Evo 2

Arc Institute / Stanford

Genome-scale foundation model spanning nucleotide to whole-genome context.

sequence generation
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

Nucleotide Transformer v2

InstaDeep / Wellcome Sanger

DNA foundation model for regulatory elements, promoters and variant scoring.

regulatory prediction
ReadyModel

ProGen2

Salesforce Research

Autoregressive protein generation up to 6.4B parameters, with an antibody variant.

sequence generation
ReadyModel

ProteinMPNN

Baker Lab

Inverse folding: given a backbone, design sequences that actually fold onto it.

inverse folding
ReadyModel

RFDiffusion

Baker Lab / Institute for Protein Design

De novo backbone generation by denoising diffusion — the binder-design workhorse.

backbone generation
ReadyModel

TxGemma-27B

Google Health AI

Open therapeutics LLM covering 66 Therapeutic Data Commons tasks.

chat

Everything else indexed

24
AbLang2Oxford Protein Informatics GroupAntibody-specific language model trained on the Observed Antibody Space.sequence embeddingOn demand
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
ANARCIOxford Protein Informatics GroupAntibody numbering and region annotation — Chothia, IMGT, Kabat.annotationOn demand
AntiFoldHummer et al., DTUInverse folding specialised for antibody backbones and CDR loops.inverse foldingOn demand
AstroLLaMAUniverseTBDLlama-2 7B fine-tuned on 300K+ astronomy arXiv abstracts for astrophysics text generation, literature summarization, and hypothesis completion — first open LLM specialized for astronomy. ================================================================================ ## Topic: Benchmark (/topics/benchmark.md) ================================================================================ # Benchmtext-generationOn 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
BioReason-Pro RLwanglabRL-tuned variant of BioReason-Pro — reinforcement-learning fine-tuning over BioReason’s SFT base for sharper biological reasoning across KEGG pathways and variant data.text-generationOn demand
BioReason-Pro SFTwanglabSupervised fine-tuned variant of BioReason-Pro — multimodal biological reasoning over genomic variants and pathway data with chain-of-thought rationales.text-generationOn demand
Carbon-3BHuggingFaceBioFlagship 3B-parameter genomic foundation model from Hugging Face Biology Research — Llama-architecture DNA language model trained for sequence generation, variant interpretation, and downstream genomic tasks.text-generationOn demand
Carbon-500MHuggingFaceBioCompact 500M-parameter Carbon variant — DNA language model lightweight enough for downstream fine-tuning on consumer hardware.text-generationOn demand
Carbon-8BHuggingFaceBioLargest 8B-parameter Carbon variant — top-of-line DNA language model for the most demanding genomic prediction, generation, and benchmark tasks.text-generationOn demand
Chai-1Chai DiscoveryOpen multimodal complex prediction across proteins, nucleic acids and ligands.complex predictionOn demand
ChatNTInstaDeepAI8B multimodal conversational model for DNA, RNA, and protein tasks — instruction-following for sequence annotation, classification, and generation.text-generationOn 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

494 more behind this view