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

Chroma

Programmable protein generation with a diffusion 'grammar'.

3/7Evidence fields documented
60-SECOND EVALUATION VIEW

What should a scientist know before using Chroma?

UnresolvedEvidence direction is incomplete or not yet resolved
Best suited forGeneration · Optimization
Evidence supportsPrimary links may be present, but BioAtlas does not claim a review date without a record-level timestamp.
Evidence does not establishUniversal superiority, therapeutic success, clinical utility or regulatory acceptance.
Major limitationPerformance depends on the evaluation dataset and operating conditions.
Current registry recordVersion history not yet curated1 recorded release · Review date not recorded. A newer version is not assumed to be universally better.

What it is

Chroma is a diffusion-based generative model for proteins that can be steered with programmable constraints — symmetry, shape, substructure and natural-language-like conditioners — to design novel proteins and complexes. It underpins Generate:Biomedicines, a Flagship Pioneering company building generative protein therapeutics.

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.

Sources3 connectedPrimary resources and normalized claims
Claims0 normalizedNo normalized claim yet
EntityChromamodel-family · Version history not yet curated
ReviewReview date not recordedReview date not claimed
ConclusionContext requiredAdd to an evaluation before operational use

Model passport

Entity typemodel-family
OrganizationGenerate Biomedicines
Model family introducedNot normalized
AccessLimited open access
Commercial useRestricted / verify terms
DeploymentHybrid
ComputeGPU recommended
Domainsdesign
Biology → representation → computation → evidence

How Chroma represents biology

model-familydesign

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

1 · Biological inputs
Target structure or design objective
2 · Input representation
Sequence and/or 3D geometry
3 · Internal representation
Generative design representation
4 · Architecture
Generative biological model
5 · Learning objective
Conditional generation
6 · Output representation
Sequence3D coordinates

Biological scale

Modalities & tasks

ProteinGenerationOptimization

Registry, claims and frontier intelligence

Versioned registry

Version history not yet curated

1 version record · release year not yet normalized. Model-family identity remains separate from capability and access changes.

Explore version lineage →
Benchmark claim ledger

0 normalized claims

No task, dataset, split and metric claim has been normalized for this record yet.

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.

Generative biomolecular design

Multimodal protein programming

EvolutionaryScale · 2025-01-16
Peer-reviewed capability

Can one generative model reason jointly over protein sequence, structure and function and create functional proteins from mixed prompts?

Evidence boundary and unresolved questions

One striking protein demonstration does not establish general success across enzymes, therapeutics or complex multi-objective design tasks.

  • How frequently do generated functions survive experimental testing?
  • Can the model optimize potency, stability and safety together?
  • How should synthetic training labels affect confidence?
multimodal · protein language model · function generation · synthetic biologyOpen frontier record →

Inputs and outputs

Inputs

Target structure or design objective

Outputs

Designed sequencesCandidate structures

Scientific and technical profile

Scientific principles

Denoising diffusionProgrammable conditioningGenerative biology

Technology

Correlated diffusionConditioners / classifiersGraph neural backbone
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.

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.

Matched concepts: diffusion, denoising, generative
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.

Matched concepts: sequence, protein
Computational intelligence

Information, entropy and communication

Claude E. Shannon

Sequence modelling, cross-entropy training, language models, mutual information and representation learning all use Shannon’s framework.

Matched concepts: sequence

Evaluation evidence

Dataset or evaluationNot yet curated
Task or metricNot yet extracted
Evidence statusPrimary paper linked; benchmark extraction pending
Open source ↗

BioAtlas has not yet extracted a structured benchmark claim for this record.

Known limitations

  • Performance depends on the evaluation dataset and operating conditions.
  • A structured benchmark claim has not yet been extracted for this record.
  • Outputs require task-specific scientific and experimental validation.

Milestones

Not normalized

Backed by Flagship Pioneering.

Evidence

Designs whole complexes, not just single chains.