# Model Documentation Evaluation Template

# Metadata
model_name: "Ministral 3 14B"
model_link: "https://legal.mistral.ai/ai-governance/models/ministral-3-14b"
organization: "Mistral AI"
org_link: "https://mistral.ai/"
evaluation_date: ""
public_summary_link: "https://legal.cms.mistral.ai/assets/36afc281-be9c-4cd0-9763-81cc19540895" 
public_summary_date: "2026-07-31"
public_summary_location: "https://legal.mistral.ai/ai-governance/models/ministral-3-14b"
model_publication_date: "2025-12-02"
category: "New model" # New model, fine-tuned model
archive_file_name: "Ministral_3_14B_2026_08_03.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: 3
    max_score: 5
    notes: ""

  D5: # Accessibility
    score: 7
    max_score: 10
    notes: ""

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

S2: # General information
  D1: # Clarity
    score: 28
    max_score: 45
    notes: ""

  D2: # Completeness
    score: 54
    max_score: 77
    notes: ""

  D3: # Consistency
    score: 8
    max_score: 8
    notes: ""

  D4: # Correctness
    score: 26
    max_score: 27
    notes: ""

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

  D6: # Comprehension
    score: 10
    max_score: 11
    notes: ""

S3: # Public datasets
  D1: # Clarity
    score: 32
    max_score: 92
    notes: ""

  D2: # Completeness
    score: 52
    max_score: 140
    notes: ""

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

  D4: # Correctness
    score: 20
    max_score: 20
    notes: ""

  D5: # Accessibility
    score: 0
    max_score: 20
    notes: ""

  D6: # Comprehension
    score: 16
    max_score: 16
    notes: ""

S4: # Private datasets
  D1: # Clarity
    score: 32
    max_score: 32
    notes: ""

  D2: # Completeness
    score: 40
    max_score: 88
    notes: ""

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

  D4: # Correctness
    score: 24
    max_score: 24
    notes: ""

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

  D6: # Comprehension
    score: 16
    max_score: 16
    notes: ""

S5: # Scraped/crawled data
  D1: # Clarity
    score: 53.5
    max_score: 114
    notes: ""

  D2: # Completeness
    score: 49
    max_score: 158
    notes: ""

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

  D4: # Correctness
    score: 53
    max_score: 93
    notes: ""

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

  D6: # Comprehension
    score: 33
    max_score: 58
    notes: ""

S6: # User data
  D1: # Clarity
    score: 0
    max_score: 15
    notes: ""

  D2: # Completeness
    score: 26
    max_score: 26
    notes: ""

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

  D4: # Correctness
    score: 11
    max_score: 11
    notes: ""

  D5: # Accessibility
    score: 15
    max_score: 15
    notes: ""

  D6: # Comprehension
    score: 0
    max_score: 15
    notes: ""

S7: # Synthetic & other
  D1: # Clarity
    score: 24
    max_score: 48
    notes: ""

  D2: # Completeness
    score: 6
    max_score: 30
    notes: ""

  D3: # Consistency
    score: 6
    max_score: 6
    notes: ""

  D4: # Correctness
    score: 30
    max_score: 30
    notes: ""

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

  D6: # Comprehension
    score: 0
    max_score: 12
    notes: ""

S8: # Data processing
  D1: # Clarity
    score: 18
    max_score: 18
    notes: ""

  D2: # Completeness
    score: 18
    max_score: 18
    notes: ""

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

  D4: # Correctness
    score: 36
    max_score: 36
    notes: ""

  D5: # Accessibility
    score: 15
    max_score: 15
    notes: ""

  D6: # Comprehension
    score: 51
    max_score: 69
    notes: ""


general_notes: "The public summary for Ministral 3 14B can be found <a href=\"https://legal.mistral.ai/ai-governance/models/ministral-3-14b\" target=\"_blank\"> on the model's section in the AI governance hub on Mistral's site</a>. The summary is largely identical to the summaries for the other Ministral 3 models. It mostly adheres to the structure of the template, however provides descriptions which do not match the degrees of detail demanded by the template. Furthermore, no list of crawled domains or full list of large public datasets is given. We assessed the summary's score to be <span class=\"openness partial\">57.14%</span> with Grade <strong>C</strong> for <strong><i>transparency</i></strong>, and <span class=\"openness partial\">58.08%</span> with Grade <strong>C</strong> for <strong><i>usefulness</i></strong>."