Provision record
Hugging Face · Hugging Face Model Card Guidelines · View original document ↗

Eval results must follow structured metadata format

High confidence Common · 267 of 343 platforms

Key Facts · in the document’s own words

How can model evaluation results be specified?
“You can specify your model's evaluation results in a structured way in the model card metadata. Results are parsed by the Hub and displayed in a widget on the model page.”
Version CA-V-006731, captured Sept. 11, 2026 · live source ↗
Our reading, not the document’s words
Hugging Face allows model evaluation results to be specified in a structured metadata format, and those results are parsed by the Hub and displayed in a widget on the model page.
Stay ahead of the changes
Track Hugging Face and hear on the next check after its terms change. We check daily where we can, and show you where we can't.
Share 𝕏 Share in Share 🔒 PDF
ⓘ

This analysis describes what Hugging Face's agreement states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability and practical outcomes may vary by jurisdiction, enforcement context, and individual circumstances. Read our methodology

ConductAtlas Analysis

Why it matters (compliance & governance perspective)

Generated by a model from the captured document; not read by a person

Because the Hub actively parses and renders this metadata, the structure and accuracy of evaluation results in the metadata directly affect what is publicly displayed on the model page.

Consumer impact (what this means for users)

Generated by a model from the captured document; not read by a person

If you add evaluation results to model card metadata in the required structured format, Hugging Face will parse and display them in a widget on your model page.

How other platforms handle this

Ford

Inform others who drive the vehicle, and passengers who connect their mobile devices to the vehicle, about the information in this Notice, including the Connected Vehicle Privacy section

Ramp

Company agrees to immediately notify Ramp in writing of any changes that may affect compliance with this Section.

Groq

Customer agrees to provide information about intended uses of the Cloud Services and AI Model Services and compliance with this policy upon request from Groq.

See all platforms with this clause type →
▸ View Original Clause Language DOCUMENT RECORD
"
You can specify your model's evaluation results in a structured way in the model card metadata. Results are parsed by the Hub and displayed in a widget on the model page.

Excerpt from Hugging Face's Model Card Guidelines

Provision details

Document information
Document
Hugging Face Model Card Guidelines
Entity
Hugging Face
Tracking information
First captured by ConductAtlas
May 12, 2026
Text quoted from version
CA-V-006731, captured Sept. 11, 2026
Record ID
CA-P-064791
Document ID
CA-D-000842
Evidence Provenance
Source URL
Wayback Machine
Archived bytes SHA-256 (version CA-V-006731)
81372a4965ee6808a6de187c5fdd1d31b0fc0b193a63f664fe05883b9e76f685
Analysis generated
July 9, 2026 06:06 UTC
Methodology
Evidence
✓ Excerpt found in version CA-V-006731, captured Sept. 11, 2026 (checked Oct. 5, 2026)
Citation Record
Entity: Hugging Face
Document: Hugging Face Model Card Guidelines
Record ID: CA-P-064791
Version: CA-V-006731
Captured: 2026-09-11 00:51:34 UTC
SHA-256: 81372a4965ee6808…
URL: https://conductatlas.com/platform/hugging-face/hugging-face-model-card-guidelines/provision/CA-P-064791/eval-results-must-follow-structured-metadata-format/
Accessed: Oct. 25, 2026
Permanent archival reference. Stable identifier suitable for legal filings, compliance documentation, and research citation.
Classification
Categories

Other risks in this policy

Get the weekly coverage letter

Companies change their terms without announcing it. We capture governance documents, compare each new version with the last, and publish what changed, with the source text behind it. One email a week on what we found and what we could not reach.

Frequently Asked Questions

What does Hugging Face's Eval results must follow structured metadata format clause do?

Because the Hub actively parses and renders this metadata, the structure and accuracy of evaluation results in the metadata directly affect what is publicly displayed on the model page.

How does this clause affect you?

If you add evaluation results to model card metadata in the required structured format, Hugging Face will parse and display them in a widget on your model page.

How many platforms have this type of clause?

ConductAtlas has identified this type of provision across 267 platforms. See the full comparison.

Is ConductAtlas affiliated with Hugging Face?

No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Hugging Face.