PARSE (Persian Zero-Shot) - Boolean Reasoning Negation: leaderboard

Metric: Accuracy (0-1; 600 yes/no reasoning questions with negation; Persian-language prompt, zero-shot, Together AI sampling). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.

Top models

#ModelScore
1Llama 3.3 70B Instruct0.82
2Qwen 2.5 72B Instruct Turbo0.81
3Gemma 2 27B (IT)0.79
4Mistral Small 30.71

Interactive version: theaggregate.ai/benchmark?slug=parse-persian-zero-shot-boolean-reasoning-negation · How It Works · Data refreshed daily, snapshot 2026-09-25.