AI for Bio · Model

RFdiffusion2 / RFdiffusion3

Atomic-level diffusion design family: RFD2 scaffolds enzyme active sites, RFD3 designs all-atom interfaces

ProteinsAcademic labUnited StatesActive中文

Overview

NameRFdiffusion2 / RFdiffusion3 · aliases: RFdiffusion2, RFdiffusion3, RFD2, RFD3
Full nameTwo models in one record, so there is no single full name
Ambiguous names (not auto-linked)RFdiffusion
FamilyCovers RFD2 and RFD3 only; the antibody fine-tuned line, the 2026-10 latent-reasoning successor branch and the sibling structure predictor are each recorded separately.
  • RFdiffusion2 · †
  • RFdiffusion3 ·
OrganizationUniversity of Washington Institute for Protein Design, Rosetta Commons
Organization typeAcademic lab
CountryUnited States
Model classDesign (diffusion, atomic level)
DomainProteins (also: Molecule generation, Structure & co-folding)
First release† partly verified · preprint
Basis: RFD2 preprint (date carried from the existing note); the RFD3 preprint 2025-09-18 and open-source release 2025-12-03 are listed separately
Note: Unverified: the first release covers RFD2 only; see the family field for the boundary source
Other milestones
  • First public evidence: † · RFD2 preprint (date carried from the existing note); the RFD3 preprint 2025-09-18 and open-source release 2025-12-03 are listed separately source
  • Preprint v1: † · RFD2 preprint (date carried from the existing note); the RFD3 preprint 2025-09-18 and open-source release 2025-12-03 are listed separately source
  • Journal online: · RFD3 journal publication (Crossref issued); the RFD2 journal version went online 2025-12-03 source
  • Code release: · RFD3 open-sourced with training code through the community Foundry source
  • Weights release: · RFD3 weights released with the open-source drop source
  • Official announcement: · Institute release page source
Latest update · journal
Basis: RFD3 journal publication (Crossref issued 2026-10-15)
RFD3's journal version appeared; the Foundry's latest tag and package release is v0.2.0 (2026-06-05) with no 1.0 tag, and commits to 2026-11-04 are untagged.
Note: The earlier 'Foundry 1.0 released' claim has no such tag and is removed source
The most recent sourced event, not a claim that nothing changed since.
StatusActive · Journal publication, a tagged release and ongoing commits within 12 months
OpennessCode: Open
Weights: Open
RFD3 ships weights and training code through a community foundation's Foundry; licence text was not retrieved this round, so the terms are left empty
Benchmarks reported1 counted (board 1)
Wet-lab validation3 recorded campaigns
Unverified fields† first_release, other_evaluations[2], open_status.code_license, open_status.weights_license
As of2026-11-16

What it does

RFD2 generates enzyme backbones conditioned on catalytic-residue geometry; RFD3 moves to a co-folding-style architecture and generates all-atom designs that interact with proteins, DNA and small molecules, with self-reported lower compute cost. Unlike the backpropagation route of BindCraft it is a trained generative model, and compared with BoltzGen it leans toward enzymes and all-atom interfaces.

Benchmarks reported

Every row is one pairing from the survey's relation file; a score belongs to the benchmark page's ranking, where it is shown with metric, eval set, source type and date. 1 of 1 pairings are counted.

BenchmarkRelationEval set / protocolSource typeNote and evidence
40 MotifBenchOfficial leaderboard or resultsofficial board entry (RFdiffusion3, 2026-08 snapshot)BoardThe board lists RFdiffusion3 (43.52, 2026-08); the Chinese model page says no entry was found, which is wrong. · evidence

Other evaluations (outside the 44)

  • Own 41-active-site set (Own task set) Self-reported: An author-built set of enzyme active sites reporting how many could be scaffolded versus the comparator (self-reported)
  • Metallohydrolase design (Wet-lab validation) Self-reported: RFD2-designed metallohydrolases were tested in two rounds and the crystal structures match the design models (journal paper)
  • Successor-branch sampling-efficiency comparison (Comparison against baselines only) Third-party: A vendor blog claims the successor branch needs fewer sampling steps and less memory; not peer reviewed and unverified

Developers report: Developers report that RFD3 beats prior methods across in silico evaluations at about one tenth of the compute cost, with wet-lab characterisation of a DNA-binding protein and a hydrolase (self-reported).

Independent evidence: The MotifBench board lists an RFD3 entry (no score quoted here).

Training data

Group in Figure 7: Public data + wet lab for validation only.

ClassValueDetail
a. Public databasesYesDistilled PDB and structure-database data source
b. Private or licensed dataNoNo private or licensed data source
c. Own wet-lab data used for trainingNoNo own training data source
d. Wet lab used only for validationYesYes: small-scale enzyme and DNA-binding-protein validation source

Wet-lab validation

Every rate carries its numerator and denominator. Claims that could not be verified are shown as unverified, without numbers; different paper versions are separate entries.

RFD2 enzyme design (self-reported) (journal version)
Targets 41 · Designs tested 96 · Hits — · Success — · Best affinity —

Denominator basis: Backbones were generated for all 41 active sites (in silico; the comparator managed 16), and for each of three catalytic mechanisms an active candidate appeared within fewer than 96 tested sequences
Verified at abstract level only, without reading full-text kinetics; scaffolding is not activity

Self-reported · † · verified · source
RFD3 wet-lab characterisation (self-reported) (preprint v2)
Targets — · Designs tested — · Hits — · Success — · Best affinity —

Denominator basis: The preprint describes characterisation of a DNA-binding protein and a cysteine hydrolase at abstract level only, with no counts
Verified at abstract level with no figures published; developer-run, not independent replication

Self-reported · † · verified · source
Successor-branch enzyme test (press release) (institutional press release)
Targets 1 · Designs tested 67 · Hits 4 · Success 4 of 67 designs showed clear activity, the best converting 79% of the substrate amine · Best affinity —

Denominator basis: 67 designs were tested; this belongs to the 2026-10 latent-reasoning successor branch, not to RFD2 or RFD3 themselves
Not peer reviewed, and the activity is catalytic conversion rather than binding; partner-lab executed, not independent replication

partner-lab · † · verified · source

Background

  • Organization: A protein design institute at a US public university, funded by several foundations, a large software company and NIH; RFD3 is released through a community foundation's open-source channel rather than run by one company.
  • Funding: Institutionally funded; no funding rounds at model level
  • Products and partnerships: It succeeds the institute's structure predictor and earlier diffusion design model; design companies spun out of the institute are not tied to this model as products.
  • Related models: RF3, RFantibody, Proteina-Complexa, BindCraft, BoltzGen

Reading notes

  1. The active-site set is author-built, and scaffolding a backbone is not the same as catalytic activity.
  2. The wet-lab volume is small and cannot support a claim that enzyme design is mature.
  3. RFD3 is open and ships training code, making it one of the few reproducible all-atom design routes.
  4. MotifBench is its only site board relation; every other evaluation sits on author-built sets.

Caveats

  1. One record covers two models, so RFD2 and RFD3 must be told apart when reading figures.
  2. The successor branch's enzyme test comes only from an institutional press release and is not peer reviewed.
  3. The earlier 'Foundry 1.0 released' claim has no such tag and is removed; the latest tag is v0.2.0 (2026-06-05).
  4. The first-release date covers RFD2 only and is marked unverified.

Sources

  1. Institute release page: RFD3 now available (official site) ·
  2. RFD3 preprint (preprint)
  3. RFD2 journal paper (journal)
  4. RFD3 journal record (Crossref) (archive)
  5. Foundry repository (tags and versions) (repository)
  6. MotifBench official repository board (leaderboard)
  7. Partner university news release (successor branch) (press)

Corrections

3 corrections
FieldWasNowBasisFound
latest_update.descriptionThe record stated that Foundry 1.0 had been releasedNo such tag exists and the claim is removed; the latest tag and package release is v0.2.0 (2026-06-05)source2026-11-16
benchmarks_reportedThe Chinese model page said no MotifBench entry was foundRFD3 is an official entry on that board (2026-08 snapshot) and the relation is official participationsource2026-11-16
release_milestones.journal_onlineThe site said RFD3 was still preprint-only as of 2026-11-16RFD3's journal version appeared 2026-10-15 and is listed as its own milestonesource2026-11-16

Part of the AI for Bio benchmarks survey · all 112 models · open in the explorer. Record as of 2026-11-16.