# Model Documentation Evaluation Template
# Metadata
model_name: "Bielik v3 11B Instruct"
model_link: "https://huggingface.co/speakleash/Bielik-11B-v3.0-Instruct"
organization: "SpeakLeash"
org_link: "https://speakleash.org/"
evaluation_date: "2026-03-30"
public_summary_link: "https://bielik.ai/downloads/Bielik%2011B%20v3%20EU%20Public%20Summary.pdf"
public_summary_date: "2026-01-01"
public_summary_location: "https://huggingface.co/speakleash/Bielik-11B-v3.0-Instruct"
model_publication_date: "2025-12-31"
category: "Fine-tuned model" # New model, fine-tuned model
archive_file_name: "Bielik_3_2026_01_12.pdf"
previous_versions:
- "2026-01-12"
# Requirement assessments
# Each requirement should be assessed once with a score from 0-10
# Leave blank for N/A, and assign a max score of 0
S1: # Document-level requirements
D1: # Clarity
score: 9
max_score: 11
notes: "Document should clearly indicate whether it is the latest version or if it is outdated and a replacement is made available.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."
D2: # Completeness
score: 4
max_score: 4
notes: ""
D3: # Consistency
score: 7
max_score: 7
notes: ""
D4: # Correctness
score: 4
max_score: 5
notes: "Document should provide link to authoritative source of the document."
D5: # Accessibility
score: 10
max_score: 10
notes: ""
D6: # Comprehension
score: 6
max_score: 9
notes: "Document should clearly indicate changes from previous version, as well as where notice of updates or changes will be provided."
S2: # General information
D1: # Clarity
score: 35.5
max_score: 44
notes: "Date of data acquisition/collection for model training should have the proper format.
Languages should be described exactly (i.e. as a list of languages rather than as the number covered)."
D2: # Completeness
score: 71
max_score: 82
notes: "Links to additional publicly available documentation should be provided for the model(s).
The EU languages which are covered should be mentioned."
D3: # Consistency
score: 8
max_score: 8
notes: ""
D4: # Correctness
score: 23
max_score: 23
notes: ""
D5: # Accessibility
score: 5
max_score: 5
notes: ""
D6: # Comprehension
score: 10
max_score: 11
notes: "No indication of whether the model is continuously trained on new or dynamic data after the date provided."
S3: # Public datasets
D1: # Clarity
score: 92
max_score: 92
notes: ""
D2: # Completeness
score: 104
max_score: 140
notes: "Public dataset field does not only provide a list of large publicly available datasets, with explanations for selecting part of the datasets where necessary.
No information relating to the types of modality contained in the datasets."
D3: # Consistency
score: 23
max_score: 23
notes: ""
D4: # Correctness
score: 20
max_score: 20
notes: ""
D5: # Accessibility
score: 20
max_score: 20
notes: ""
D6: # Comprehension
score: 16
max_score: 16
notes: ""
S4: # Private datasets
D1: # Clarity
score: 0
max_score: 0
notes: ""
D2: # Completeness
score: 5
max_score: 5
notes: ""
D3: # Consistency
score: 3
max_score: 3
notes: ""
D4: # Correctness
score: 5
max_score: 5
notes: ""
D5: # Accessibility
score: 0
max_score: 0
notes: ""
D6: # Comprehension
score: 0
max_score: 0
notes: ""
S5: # Scraped/crawled data
D1: # Clarity
score: 109
max_score: 119
notes: "Field describing purpose of crawlers does not only contain information relating to the purposes of the crawlers listed."
D2: # Completeness
score: 110
max_score: 135
notes: "No list of the most relevant internet domains is provided, as per the requirements."
D3: # Consistency
score: 3
max_score: 3
notes: ""
D4: # Correctness
score: 88
max_score: 113
notes: "No list of the most relevant internet domains is provided, as per the requirements."
D5: # Accessibility
score: 0
max_score: 25
notes: "No list of the most relevant internet domains is provided, as per the requirements."
D6: # Comprehension
score: 53
max_score: 78
notes: "No list of the most relevant internet domains is provided, as per the requirements."
S6: # User data
D1: # Clarity
score: 0
max_score: 0
notes: ""
D2: # Completeness
score: 8
max_score: 8
notes: ""
D3: # Consistency
score: 3
max_score: 3
notes: ""
D4: # Correctness
score: 8
max_score: 8
notes: ""
D5: # Accessibility
score: 0
max_score: 0
notes: ""
D6: # Comprehension
score: 0
max_score: 0
notes: ""
S7: # Synthetic & other
D1: # Clarity
score: 24
max_score: 24
notes: ""
D2: # Completeness
score: 18
max_score: 18
notes: ""
D3: # Consistency
score: 6
max_score: 6
notes: ""
D4: # Correctness
score: 18
max_score: 18
notes: ""
D5: # Accessibility
score: 0
max_score: 0
notes: ""
D6: # Comprehension
score: 12
max_score: 12
notes: ""
S8: # Data processing
D1: # Clarity
score: 18
max_score: 19
notes: "Information in 3.3 is not all relevant to data processing aspects and measures taken before or after training model."
D2: # Completeness
score: 19
max_score: 19
notes: ""
D3: # Consistency
score: 3
max_score: 3
notes: ""
D4: # Correctness
score: 37
max_score: 37
notes: ""
D5: # Accessibility
score: 0
max_score: 15
notes: "Not all information relating to the description of removal of illegal content is pertinent."
D6: # Comprehension
score: 60
max_score: 69
notes: "Provider does not describe how they ensure that they are up to date with user rights requests."
general_notes: "The open collaboration project SpeakLeash publishes a public summary for its Bielik v3 11B Instruct model on its website and also links it in its model card. The public summary follows the structure of the template, and describes Bielik as a fine-tuned version of Mistral. Of note, Bielik utilised the optional additional comments field in Section 1.3 to describe the purposes, regional, and linguistic characteristics of the dataset used as based on Polish and EU legal administrative domains and the inclusion of Silesian and Kashubian language Wikipedia pages. For their list of public data sources, however, the description referenced external documents ('preprints') which resulted in a deduction of points as relevant information was not provided in the document. We assessed its score to be 88.02% with grade B+ for transparency, and 71.11% with grade C+ for usefulness."