SmolLM3-3B

GPAI Provider: HuggingFace

Metadata

Model URL: https://huggingface.co/HuggingFaceTB/SmolLM3-3B
When was this model published? 2025-07-08
Is this a new or a fine-tuned model? Fine-tuned model
When was the Public Summary last updated? 2025-07-25
When did we evaluate the Public Summary? 2026-03-31 (see previous versions here)
Where did we find the Public Summary? We found this document linked on this page. An archived version can be found here.

Results of Evaluation

Section Transparency Usefulness General
Clarity Completeness Consistency Correctness Accessibility Comprehension Transparency Usefulness
Document 68.18 100.0 68.75 50.0 65.38 50.0 68.97 58.0
General information 52.5 83.08 100.0 73.91 N/A 90.91 73.53 90.91
Public Data Sources 84.62 100.0 100.0 100.0 100.0 100.0 96.06 100.0
Private Data Sources N/A 100.0 100.0 100.0 N/A N/A 100.0 N/A
Scraped/Crawled Data N/A 100.0 N/A 100.0 N/A N/A 100.0 N/A
User Data N/A 100.0 N/A 100.0 N/A N/A 100.0 N/A
Synthetic/Other Data 67.57 61.29 100.0 41.94 N/A 0.0 60.0 0.0
Data Processing 100.0 83.33 100.0 91.67 100.0 100.0 92.0 100.0
Sum 73.1 89.12 95.1 76.74 90.62 84.17 82.5 86.01
Grades B+B+

Evaluation Notes

If you update your public summary, please let us know so that we can update the evaluation and scores. We also welcome suggestions on improvements to this page, e.g. see FAQ section.

The public summary for HuggingFace's SmolLM3-3B model is available on its platform as a webpage. In general, the structure of the template was preserved with clear indication of section numbers, consistent field labels, and explicitly indicating yes/no where applicable. Other than issues with accessibility, we found no major issues with the public summary. It scored 82.5% and received a grade of B+ for Transparency, and 86.01% with grade B+ for Usefulness, which meant that the summary was of a moderate but good quality. SmolLM scored particularly well on its disclosure of public data sources, a key disclosure demanded by Article 53(1)(d), and a good indicator of transparency in the model development process.

Suggested Improvements
The following are the suggested improvements based on using our our methodology where the public summary had issues related to the specified metric. The severity represents the extent of the issue, with low indicating aspects that could be fixable, and high representing missing information or requiring major changes.

  • Document
    • M1 Document should have the correct structure, and be in a well-defined, structured, and interoperable format.

      Document should clearly indicate whether it is the latest version or if it is outdated and a replacement is made available.

      Document should provide link to all versions of the document. medium
    • M3 Document should be consistent across versions. Each document version should have a unique version number. medium
    • M4 Document should have assured integrity and provide a link to the authoritative source of the document. The date of last update should be correct and accurate. medium
    • M5 Document must be easy to find, support sharing and exporting, and be provided in the same context as the model. medium
    • M6 Document should clearly indicate changes from previous version, or otherwise where notice of updates or changes will be provided. Links to previous versions should be accessible for intended stakeholders. medium
  • General information
    • M1 Contact details refer to a generic contact point. Description of types of content contains some description pertaining to goals rather than types.

      Date of latest acquisition/collection for model training should have the proper format. Not all included languages are provided as a list. medium
    • M2 Model dependencies should include SmolLM3-3B-Base. Date of placement on the Union market should be indicated. low
    • M4 Model dependencies should include SmolLM3-3B-Base. Date of placement on the Union market should be indicated. medium
    • M6 No indication of whether the model is continuously trained on new or dynamic data after the date provided. low
  • Public Data Sources
    • M1 General description of other publicly available datasets contains some superfluous information. low
  • Synthetic/Other Data
    • M1 No information about other AI models used to generate synthetic data to train the model to which this Summary applies. medium
    • M2 No information about other AI models used to generate synthetic data to train the model to which this Summary applies. medium
    • M4 Slight inconsistency in general-purpose AI model(s) used to generate the synthetic data if available on the market. No information about other AI models used to generate synthetic data to train the model to which this Summary applies. high
    • M6 No information about other AI models used to generate synthetic data to train the model to which this Summary applies. high
  • Data Processing
    • M2 Signatory to the Code of Practice field not filled in. low
    • M4 Signatory to the Code of Practice field not filled in. low
  • Sum
    • M1 Document should have the correct structure, and be in a well-defined, structured, and interoperable format.

      Document should clearly indicate whether it is the latest version or if it is outdated and a replacement is made available.

      Document should provide link to all versions of the document.Contact details refer to a generic contact point. Description of types of content contains some description pertaining to goals rather than types.

      Date of latest acquisition/collection for model training should have the proper format. Not all included languages are provided as a list.General description of other publicly available datasets contains some superfluous information.No information about other AI models used to generate synthetic data to train the model to which this Summary applies. medium
    • M2 Model dependencies should include SmolLM3-3B-Base. Date of placement on the Union market should be indicated.No information about other AI models used to generate synthetic data to train the model to which this Summary applies.Signatory to the Code of Practice field not filled in. low
    • M3 Document should be consistent across versions. Each document version should have a unique version number. low
    • M4 Document should have assured integrity and provide a link to the authoritative source of the document. The date of last update should be correct and accurate.Model dependencies should include SmolLM3-3B-Base. Date of placement on the Union market should be indicated.Slight inconsistency in general-purpose AI model(s) used to generate the synthetic data if available on the market. No information about other AI models used to generate synthetic data to train the model to which this Summary applies.Signatory to the Code of Practice field not filled in. low
    • M5 Document must be easy to find, support sharing and exporting, and be provided in the same context as the model. low
    • M6 Document should clearly indicate changes from previous version, or otherwise where notice of updates or changes will be provided. Links to previous versions should be accessible for intended stakeholders.No indication of whether the model is continuously trained on new or dynamic data after the date provided.No information about other AI models used to generate synthetic data to train the model to which this Summary applies. low

Previous Versions