Skip to main content
model-family passport · Review date not recorded

Boltz-1 / Boltz-2

Open-source AF3-quality structure — plus binding affinity.

4/7Evidence fields documented
60-SECOND EVALUATION VIEW

What should a scientist know before using Boltz-1 / Boltz-2?

SupportedEvidence supports the stated context with explicit boundaries
Best suited forPrediction · Affinity prediction
Evidence supportsPoseBusters and affinity benchmarks: Preprint / open evaluation
Evidence does not establishUniversal superiority, therapeutic success, clinical utility or regulatory acceptance.
Major limitationPerformance depends on the evaluation dataset and operating conditions.
Current registry recordBoltz-22 recorded releases · Review date not recorded. A newer version is not assumed to be universally better.

What it is

Boltz-1 was the first fully open, commercially-usable model to match AlphaFold 3 on biomolecular complex prediction. Boltz-2 (2025) added fast, accurate binding-affinity prediction — approaching physics-based free-energy methods at a fraction of the cost — a key missing piece for virtual screening.

Evidence trail

BioAtlas keeps the path from source to decision visible. A connection records provenance; it does not imply that evidence is sufficient for every context.

Sources4 connectedPrimary resources and normalized claims
Claims1 normalizedBiomolecular complex prediction
EntityBoltz-1 / Boltz-2model-family · Boltz-2
ReviewReview date not recordedReview date not claimed
ConclusionContext requiredAdd to an evaluation before operational use

Model passport

Entity typemodel-family
OrganizationMIT (Barzilay & Jaakkola labs)
Model family introduced2025
AccessOpen source
Commercial useAllowed / verify checkpoint terms
DeploymentSelf-hosted
ComputeGPU recommended
Domainsstructure · chemistry
Biology → representation → computation → evidence

How Boltz-1 / Boltz-2 represents biology

model-familystructurechemistry

Category is navigation. These fields describe the model-specific computational transformation and deliberately override broad category defaults.

1 · Biological inputs
Biomolecular complex specification
2 · Input representation
Sequence / molecular tokensAll-atom context
3 · Internal representation
Geometric pair representations
4 · Architecture
AF3-class diffusion modelAffinity head (Boltz-2)
5 · Learning objective
Complex coordinate predictionBinding-affinity prediction
6 · Output representation
All-atom coordinatesAffinity scores

Biological scale

proteincomplexmolecule

Modalities & tasks

ProteinDNARNALigandPredictionAffinity prediction

Registry, claims and frontier intelligence

Versioned registry

Boltz-2

2 version records · latest curated year 2025. Model-family identity remains separate from capability and access changes.

Explore version lineage →
Benchmark claim ledger

1 normalized claim

Biomolecular complex prediction · PoseBusters and affinity benchmarks

Open claim intelligence →

Connected research frontiers

These records describe active research directions, not guaranteed capabilities of this model. Evidence stages and unresolved questions are preserved separately.

Dynamic structure & docking

Open reproductions of frontier drug-design engines

Aureka AI OpenDDE project · 2026-07-04
Recent preprint

Can the community reproduce and extend proprietary all-atom drug-design engines with open training code, checkpoints and benchmarks?

Evidence boundary and unresolved questions

OpenDDE is a very recent July 2026 preprint. Its claimed parity has not yet received broad independent evaluation.

  • Can external teams reproduce the reported training and benchmark results?
  • What data provenance and leakage controls are documented?
  • How do open checkpoints perform in prospective discovery projects?
open science · co-folding · reproducibility · scaling lawsOpen frontier record →
Affinity & virtual screening

Joint structure and binding-affinity reasoning

Isomorphic Labs · MIT · Open community · 2026-03-06
Recent preprint

Can one model predict both the bound geometry and the energetic ranking needed to prioritize compounds?

Evidence boundary and unresolved questions

Recent independent evaluations report that strong global correlations may not preserve ranking among top compounds, where lead-selection decisions occur.

  • Can models rank close analogues within a chemical series?
  • How well do they extrapolate across assays, protonation states and target families?
  • Can uncertainty identify when physics or experiment should take over?
binding affinity · ranking · virtual screening · free energy · uncertaintyOpen frontier record →
Dynamic structure & docking

Finding cryptic pockets from sequence

Isomorphic Labs · 2026-02-10
Developer-reported

Can a model reveal ligandable pockets that are hidden in the unbound protein and only open after a ligand or allosteric change?

Evidence boundary and unresolved questions

The public evidence is currently a company technical report and benchmark narrative. Prospective medicinal-chemistry validation and independent replication remain essential.

  • How often are predicted pockets experimentally ligandable rather than geometrically plausible?
  • How robust is pocket discovery across membrane proteins, intrinsically disordered regions and low-data target families?
  • Can calibrated confidence distinguish genuine induced pockets from model hallucinations?
cryptic pockets · allostery · induced fit · ligandability · dockingOpen frontier record →
Dynamic structure & docking

Induced-fit co-folding beyond familiar targets

Isomorphic Labs · 2026-02-10
Developer-reported

Can structure models represent large ligand-driven protein rearrangements when the target, pocket or conformational transition is far from training examples?

Evidence boundary and unresolved questions

Out-of-distribution claims depend strongly on benchmark construction, training-set leakage controls and exact success thresholds.

  • How are unseen chemotypes and target families isolated from training data?
  • Does structural accuracy translate into enrichment or medicinal-chemistry decisions?
  • How stable are alternative conformational ensembles?
co-folding · induced fit · OOD generalization · protein flexibilityOpen frontier record →
Generative biomolecular design

High-fidelity antibody and biologic interfaces

Isomorphic Labs · 2026-02-10
Developer-reported

Can general co-folding models accurately resolve antibody–antigen and other biologic interfaces with low sequence homology?

Evidence boundary and unresolved questions

Benchmark composition, success thresholds and independent reproduction will determine how broadly the reported advantage generalizes.

  • How does performance vary across CDR loops, nanobodies and multispecific formats?
  • Can interface prediction improve prospective affinity maturation?
  • How are glycosylation and conformational heterogeneity handled?
antibodies · biologics · interfaces · low homologyOpen frontier record →
Affinity & virtual screening

Fast, uncertainty-aware affinity screening

Terray Therapeutics research team · 2026-02-08
Recent preprint

Can virtual screening retain useful structural and affinity accuracy without expensive all-atom diffusion for every compound?

Evidence boundary and unresolved questions

The results are preprint claims; proprietary assay details and cross-lab prospective replication remain limited.

  • Does coarse representation preserve water, ion, metal and covalent chemistry?
  • How well does uncertainty calibrate under target and chemistry shift?
  • Can active learning improve real design-make-test cycles?
virtual screening · uncertainty · active learning · coarse representationOpen frontier record →

Inputs and outputs

Inputs

Biomolecular complex specification

Outputs

Complex structuresBinding-affinity predictions

Scientific and technical profile

Scientific principles

Diffusion generative structureAffinity predictionOpen science

Technology

AF3-style diffusionAffinity moduleMIT-licensed weights
Ideas before algorithms

Scientific lineage

Explore all foundations

These are transparent concept matches—not claims that one scientist alone caused this model. Each connection is based on the model’s recorded domain, scientific principles, technical terms or an explicit lineage link.

Structural biology

Anfinsen’s dogma—the thermodynamic hypothesis

Christian B. Anfinsen

Protein structure prediction, inverse folding and generative protein design all assume that sequence strongly constrains structure and function.

Explicit model lineage
Computational intelligence

Denoising diffusion generative models

Jascha Sohl-Dickstein, Jonathan Ho and collaborators

Modern protein-backbone, molecular-pose and biomolecular-complex generators use diffusion to sample valid three-dimensional structures and designs.

Explicit model lineage
Structural biology

First atomic structures of globular proteins

John Kendrew & Max Perutz

Protein structure prediction and structure-based design became meaningful because experimental crystallography established the target reality to predict against.

Matched concepts: fold, complex
Structural biology

Levinthal’s paradox and efficient folding pathways

Cyrus Levinthal

Modern folding algorithms, energy landscapes, learned priors and diffusion models solve a constrained search problem rather than brute-force conformational enumeration.

Matched concepts: fold, diffusion
Physical chemistry

Multiscale modelling of chemical systems

Martin Karplus, Michael Levitt & Arieh Warshel

QM/MM, molecular dynamics, free-energy methods and physics–ML hybrid platforms descend directly from this multiscale strategy.

Matched concepts: physics, free-energy

Evaluation evidence

Dataset or evaluationPoseBusters and affinity benchmarks
Task or metricStructure and affinity
Evidence statusPreprint / open evaluation
Open source ↗

Task-specific evidence only; not comparable as a universal leaderboard score.

Biomolecular complex prediction

PoseBusters and affinity benchmarks

Boltz-2 · PoseBusters and affinity evaluation sets
developer-reported

Open evaluation reports structure prediction and later affinity capabilities.

Claim caveats
  • Boltz-1 structure claims and Boltz-2 affinity claims should be separated by version.
  • Affinity performance depends strongly on target family and split design.

Known limitations

  • Performance depends on the evaluation dataset and operating conditions.
  • Task-specific benchmark results should not be compared across unlike domains.
  • Outputs require task-specific scientific and experimental validation.

Milestones

2025

BoltzGen extends the line toward universal binder design.

Evidence

Widely used as an open AF3 alternative.