Political-bias benchmark for Grok 4, GPT-4.1, Gemini 2.5 Pro and Claude Opus 4

3 points| dangelosaurus | 7 months ago |promptfoo.dev