Independent project. Not a U.S. government website.

USASI
Model family

Boltz-2

Family overview — summarizes releases; licenses and availability belong to each release.

Maintained by Boltz (Boltz PBC)671

Boltz-2 is a biomolecular foundation model that predicts the 3D structures of complexes of proteins, nucleic acids, and small molecules together with protein-ligand binding affinity. It was released in June 2025 by the Boltz team at MIT Jameel Clinic with Recursion and is now presented by Boltz PBC as one of its open-source models. It ships as a structure and confidence checkpoint plus a separate affinity checkpoint.463Fact reviewed Oct 1, 2026

Last reviewedEntry updated Documented release Jun 6, 2025

Releases assessed

Each release is assessed on its own. This family has no family-wide openness label.
Releases in the Boltz-2 family
ReleaseWeightsTier (USASI rubric v0.2)LicenseReleased
Boltz-2 (boltz2_conf and boltz2_aff checkpoints)Weights: PublicModel-disclosure tier (USASI rubric v0.2): Open-weightMITJun 6, 2025

What it is useful for

The README describes structure prediction from YAML inputs and two affinity outputs: a binder probability intended for separating binders from decoys in hit discovery, and an affinity value intended for comparing related binders during hit-to-lead and lead optimization. The announcement adds contact constraints, template guidance, and method-conditioning options.14Fact reviewed Oct 1, 2026

Organization context

U.S. eligibility

Project eligibility rests on documented governing or maintaining entities, not on contributors.

Eligible · basis: U.S.-governed project

Boltz-2 was developed by the Boltz team at MIT Jameel Clinic with Recursion, and Boltz PBC, whose terms give a Massachusetts notice address, now presents it as one of its open-source models and links the repository from its website. The technical report also lists authors at Valence Labs and ETH Zurich. The code repository sits under an individual developer's GitHub account, and its license names three individual copyright holders.46825

Assessed Oct 1, 2026

Sources

  1. 1.
    jwohlwend/boltz README (external site: raw.githubusercontent.com)

    Boltz developers (GitHub) · Repository · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  2. 2.
    jwohlwend/boltz LICENSE (external site: raw.githubusercontent.com)

    Boltz developers (GitHub) · License · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  3. 3.
    boltz-community/boltz-2 (Hugging Face model API metadata) (external site: huggingface.co)

    Hugging Face · Other · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  4. 4.
    Introducing Boltz-2: Toward Accurate and Efficient Binding Affinity Prediction (external site: boltz.com)

    Boltz · Announcement · published Jun 6, 2025 · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  5. 5.
    Boltz-2: Towards Accurate and Efficient Binding Affinity Prediction (bioRxiv v1) (external site: biorxiv.org)

    bioRxiv · Paper · published 2025 · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  6. 6.
    Boltz - Build Better Molecules with AI (home page) (external site: boltz.com)

    Boltz · Official page · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  7. 7.
    The future we are building at Boltz (external site: boltz.com)

    Boltz · Announcement · published Jan 8, 2026 · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

  8. 8.
    Boltz Terms of Service (external site: boltz.com)

    Boltz PBC · Official page · published May 6, 2026 · accessed Oct 1, 2026 · evidence reviewed Oct 1, 2026

Support Us

Help keep USASI useful.

Optional. No USASI account required. Payment takes place on the linked provider’s website (Buy Me a Coffee).

About supporting this project