ESMFold2 Paper and bioRxiv Status

ESMFold2 has paper and preprint search intent because users want the primary scientific source behind the Biohub release. This page helps users find the current source path without inventing a DOI, title variant, or benchmark number.

Before citing the work, open the Biohub model page and the Biohub ESM GitHub repository. If those sources link a paper, PDF, or preprint, cite that current target directly.

Verify the paper title, author list, publication venue, date, DOI, and whether the link is a Biohub-hosted PDF, preprint server entry, journal article, or repository-linked technical report.

If a bioRxiv DOI is not visible in the current primary source path, do not write one into a page, schema object, or generated citation. Keep the wording to paper or preprint status until the DOI is verified.

This page keeps citation intent separate from implementation intent. A user checking the paper should verify the paper link, DOI visibility, venue, authors, and date before reusing a title, benchmark claim, or schema citation.

When users need runnable details, the better next step is the GitHub or API guide. When they need to inspect an output artifact, the demo result page is a safer destination than treating a paper page as an execution tutorial.

Scientific claims should stay tied to the source that makes them. A benchmark statement from Biohub should be written as a Biohub-reported claim; a third-party article should be named as reporting context.

This page should not turn selected evaluation language into a broad claim that ESMFold2 is universally better than another model. It should route readers to primary sources for exact datasets and limitations.

The paper is best for scientific motivation, model framing, method details, and evaluation context. The GitHub repository and official notebook are better for current runnable examples.

That separation matters because implementation details can change faster than paper text. When a class name, model ID, package path, or notebook cell matters, verify it against current official docs before running or publishing code.

Start from the Biohub model page, follow its paper link, then check the Biohub ESM repository for the same reference. If Nature or another third-party source summarizes the release, use it as reporting context rather than as the primary implementation source.

Keep a visible last-updated date on the page. If the paper moves from a Biohub-hosted PDF to a preprint server or journal page, update the source link and date together.

This page does not claim to host the official paper, does not publish independent benchmark results, and does not replace Biohub documentation.

It also does not promise that older ESMFold paper citations apply to ESMFold2. Older ESMFold literature can provide background, but ESMFold2 claims should be checked against the Biohub release and the current paper source.

This launch page is written as a third-party guide, not as an official Biohub page. It links to Biohub, GitHub, Colab, model-card, paper, and reporting sources so readers can verify current details before copying code or making technical decisions.

The free Track A site does not collect protein sequences, private Biohub tokens, prediction outputs, or uploaded structure files. Browser-local tools can prepare starter code and examples, but hosted prediction, stored results, queues, and credits remain Phase 2 features.

Predicted structures should not be described as experimental validation, clinical evidence, or drug-development proof. When this page mentions model scope or comparisons, it keeps claims tied to official or clearly attributed sources.

If a source changes after this page is published, the source should win. The page should be updated with a new verification date instead of preserving stale client names, model IDs, pricing assumptions, or unsupported benchmark language.