Skip to main content
MODERATION BIAS
AI OverviewLeaderboardComparisonModelsCategoriesAnnotatePrompts
SummaryReliabilityLongitudinal AnalysisModel StabilitySignificanceAnnotator AgreementFamily AnalysisPolitical CompassPaternalismAlignment TaxOver-Refusal
Semantic ClustersTrigger ListCouncil Consensus
AboutMethodologyGlossary

Cite This Research

BibTeX
@misc{kandel2026moderationbias,
  title     = {Moderation Bias: A Systematic Benchmark of Content Moderation Across Large Language Models},
  author    = {Kandel, Jacob},
  year      = {2026},
  url       = {https://moderationbias.com},
  note      = {Open benchmark and dataset available at https://huggingface.co/datasets/jmk9494/moderation-bias-benchmark}
}
APA

Kandel, J. (2026). Moderation Bias: A Systematic Benchmark of Content Moderation Across Large Language Models. https://moderationbias.com

  1. Models
  2. Mistralai

Mistral AI

Models provided by Mistral AI.

Mistral Large logo

Mistral Large

High

Mistral Small 3.1 logo

Mistral Small 3.1

Mid

Mistral Small 3 logo

Mistral Small 3

Mid

Ministral 8B logo

Ministral 8B

Low

Mistral Small 3.1 logo

Mistral Small 3.1

Low

© 2026 Moderation Bias. All rights reserved.