Evaluated 6 frontier LLMs (GPT-5.4, Claude Sonnet 4.6, Claude Opus 4.7, Gemini Pro/Flash, Grok 4.3) on political, gender, and racial bias across 8 benchmarks (~20,600 examples) [R]

I ran a solo evaluation project benchmarking six current frontier models: GPT-5.4, Claude Sonnet 4.6, Claude Opus 4.7, Gemini Pro, Gemini Flash, and Grok 4.3. I tested tham across 8 established bias/fairness datasets (WinoBias, BBQ Race/Ethnicity, SeeGULL, OpinionsQA, cajcodes Political Bias, Hyperp