rewirebio.iobenchmarks
Model

mRNA-FM

mRNA-FM encodes coding RNA with codon-level tokens to produce representations for downstream analysis.

Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example

Results are recorded on linked configurations and versions of this model. Their settings and evaluations are kept separate below.

1 pipeline with results

How it worksmRNA-FM workflow
mRNA-FM workflow1. Coding RNA. Then: 2. Codon tokenizer. Then: 3. 12-layer transformer encoder. Then: 4. Contextual codon representationsmRNA-FM workflow1. Coding RNA. Then: 2. Codon tokenizer. Then: 3. 12-layer transformer encoder. Then: 4. Contextual codon representationsmRNA-FM workflow1. Coding RNA. Then: 2. Codon tokenizer. Then: 3. 12-layer transformer encoder. Then: 4. Contextual codon representations

Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.

Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example

Results on linked records

This model has no results of its own in this release. One linked record has results. They are not assigned to the underlying model: each record keeps its own settings and evaluations, and its scores are on its own page.

  • mrna-fm Pipeline · uses this model · 8 evaluations · 8 results

Overview

Model type

Codon-token RNA transformer encoder

Inputs

Coding RNA in the correct reading frame, with sequence length divisible by three.

Outputs

Contextual token embeddings for a specified downstream RNA task.

Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example

limited source coverage · Automated source review, 2026-09-16. All specifications and missing details

Model Deployment Profile

mRNA-FM has no ranked result on a use-case protocol in this release.

Recorded parameters: 239M, as printed in the official Foundation Models table.. Code licence: MIT.

Model Capabilities & Hardware Requirements

Deployment facts recorded for mRNA-FM. Fields the record does not carry are marked not recorded.

Parameters239M, as printed in the official Foundation Models table
Minimum VRAMNot recorded
Inference LatencyNot recorded
Context WindowNot recorded
AccessPublic checkpoint trained on coding sequences (CDS); input…
Code licenceMIT

Evaluated Checkpoint: codon-tokenised

Use this model

How it works, versions and access

Related profile: RNA-FM. This page retains the exact record and its evaluation context.

How it works

How it works

12-layer transformer encoder with hidden width 1,280 and codon-level input tokens. The documented inputs are coding RNA in the correct reading frame, with sequence length divisible by three. The output consists of contextual token embeddings for a specified downstream RNA task.

Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Versions and reproducibility

mrna_fm_t12; distinct from base-tokenized rna_fm_t12. The applicable input limits require configuration-specific checking.

Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Strengths, limitations and unresolved questions

Strengths and limitations

Limitations and conditions

Profile review details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Stable record: catalog-model-mrna-fm

Specifications

Inputs, training, access and other details

Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Inputs, outputs and configuration
PropertyDescription and evidence
Model typeCodon-token RNA transformer encoder
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Architecture12-layer transformer encoder with hidden width 1,280 and codon-level input tokens.
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
InputsCoding RNA in the correct reading frame, with sequence length divisible by three.
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
OutputsContextual token embeddings for a specified downstream RNA task.
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Parameters239M, as printed in the official Foundation Models table.
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Known versionsmrna_fm_t12; distinct from base-tokenized rna_fm_t12.
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Training data45M messenger RNA sequences, as printed in the official Foundation Models table.
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Training cutoffThe official model table reports 45M coding RNAs but does not identify a latest-data date for that separate mRNA-FM corpus. · Not reported in inspected sources
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Context limitsThe inspected mRNA-FM usage example and loader do not specify a validated maximum codon-sequence length. The retrieved original RNA-FM paper describes the nucleotide-based non-coding RNA model; its limit cannot establish the separate mRNA-FM checkpoint limit. · Not reported in inspected sources
Sources (3)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py; RNA-FM original methods, arXiv v5 · Official README: Foundation Models and mRNA-FM quick start; loader; original RNA-FM paper Methods, pp.22–23
Weights licenceSeparate checkpoint-distribution terms are not stated in the inspected release documentation and licence material. The source-code licence alone is not recorded as an explicit weight grant. · Not reported in inspected sources
Sources (3)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py; ml4bio/RNA-FM: LICENSE · README.md: Foundation Models table and mRNA-FM quick-start example; LICENSE: licence text
AccessOfficial project documentation and implementation: https://github.com/ml4bio/RNA-FM
Sources (2)ml4bio/RNA-FM: README.md; ml4bio/RNA-FM: fm/pretrained.py · README.md: Foundation Models table and mRNA-FM quick-start example
Code licenceMIT
Sourcesml4bio/RNA-FM: LICENSE · LICENSE: licence text

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

45 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-c8bea25e5abb
Property and statementOriginal source and locationReview and provenance
Diagram caption
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Individual claims
ml4bio/RNA-FM: fm/pretrained.py

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.diagram.caption

Source artifact SHA-256: ab7271f8dbb876dbc5e2010e2d88b086088aada096f04149436d315a4f256aac

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram caption
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Individual claims
ml4bio/RNA-FM: README.md

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.diagram.caption

Source artifact SHA-256: f9f1c1d62adc471661ca98b30c0250e9f3ce0cff7433830f149f5f48ea41c3da

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram steps
  • Coding RNA
  • Codon tokenizer
  • 12-layer transformer encoder
  • Contextual codon representations
Individual claims
ml4bio/RNA-FM: fm/pretrained.py

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.diagram.steps

Source artifact SHA-256: ab7271f8dbb876dbc5e2010e2d88b086088aada096f04149436d315a4f256aac

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram steps
  • Coding RNA
  • Codon tokenizer
  • 12-layer transformer encoder
  • Contextual codon representations
Individual claims
ml4bio/RNA-FM: README.md

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.diagram.steps

Source artifact SHA-256: f9f1c1d62adc471661ca98b30c0250e9f3ce0cff7433830f149f5f48ea41c3da

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram title
mRNA-FM workflow
Individual claims
ml4bio/RNA-FM: fm/pretrained.py

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.diagram.title

Source artifact SHA-256: ab7271f8dbb876dbc5e2010e2d88b086088aada096f04149436d315a4f256aac

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram title
mRNA-FM workflow
Individual claims
ml4bio/RNA-FM: README.md

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.diagram.title

Source artifact SHA-256: f9f1c1d62adc471661ca98b30c0250e9f3ce0cff7433830f149f5f48ea41c3da

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Model type
Codon-token RNA transformer encoder
Individual claims
ml4bio/RNA-FM: fm/pretrained.py

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.facts.0.value

Source artifact SHA-256: ab7271f8dbb876dbc5e2010e2d88b086088aada096f04149436d315a4f256aac

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Model type
Codon-token RNA transformer encoder
Individual claims
ml4bio/RNA-FM: README.md

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.facts.0.value

Source artifact SHA-256: f9f1c1d62adc471661ca98b30c0250e9f3ce0cff7433830f149f5f48ea41c3da

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Architecture
12-layer transformer encoder with hidden width 1,280 and codon-level input tokens.
Individual claims
ml4bio/RNA-FM: fm/pretrained.py

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.facts.1.value

Source artifact SHA-256: ab7271f8dbb876dbc5e2010e2d88b086088aada096f04149436d315a4f256aac

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Architecture
12-layer transformer encoder with hidden width 1,280 and codon-level input tokens.
Individual claims
ml4bio/RNA-FM: README.md

Original source ↗

README.md: Foundation Models table and mRNA-FM quick-start example

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 348951516e0963d22bbb33b3c9fc18c89081d38e
Retrieved: 2026-09-16T19:46:19.769679+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Follow-up retrieved the complete original RNA-FM PDF and inspected the full pinned Geneformer repository inventory; unavailable labels were updated only where new evidence resolved the earlier retrieval gap.

Field: attributes.profile.facts.1.value

Source artifact SHA-256: f9f1c1d62adc471661ca98b30c0250e9f3ce0cff7433830f149f5f48ea41c3da

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

How to read the evidence

Sources and history

Release 2026-11-21-c8bea25e5abb · Record review: discovered

5 source records and release historyDownload this release (gzip)
Technical metadata and record identifiers

Technical metadata for this record is tracked under stable ID catalog-model-mrna-fm.

Related records

How to cite

mRNA-FM. Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/model/catalog-model-mrna-fm/

Release 2026-11-21-c8bea25e5abb. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.

The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence

Suggest a correction