# 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: "2026-01-12"
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"

# 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: ""

  D2: # Completeness
    score: 4
    max_score: 4
    notes: ""

  D3: # Consistency
    score: 7
    max_score: 7
    notes: ""

  D4: # Correctness
    score: 4
    max_score: 5
    notes: ""

  D5: # Accessibility
    score: 10
    max_score: 10
    notes: ""

  D6: # Comprehension
    score: 6
    max_score: 9
    notes: ""

S2: # General information
  D1: # Clarity
    score: 35.5
    max_score: 45
    notes: ""

  D2: # Completeness
    score: 71
    max_score: 82
    notes: ""

  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: ""

S3: # Public datasets
  D1: # Clarity
    score: 92
    max_score: 92
    notes: ""

  D2: # Completeness
    score: 104
    max_score: 140
    notes: ""

  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: ""

  D2: # Completeness
    score: 110
    max_score: 135
    notes: ""

  D3: # Consistency
    score: 3
    max_score: 3
    notes: ""

  D4: # Correctness
    score: 88
    max_score: 113
    notes: ""

  D5: # Accessibility
    score: 0
    max_score: 25
    notes: ""

  D6: # Comprehension
    score: 53
    max_score: 78
    notes: ""

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: ""

  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: ""

  D6: # Comprehension
    score: 60
    max_score: 69
    notes: ""


general_notes: "The open collaboration project SpeakLeash publishes a public summary for its Bielik v3 11B Instruct model <a href=\"https://bielik.ai/downloads/Bielik%2011B%20v3%20EU%20Public%20Summary.pdf\" target=\"_blank\">on its website</a> 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 <span class=\"openness open\">88.02%</span> with grade <strong>B+</strong> for <i>transparency</i>, and <span class=\"openness partial\">71.11%</span> with grade <strong>C+</strong> for <i>usefulness</i>."