PARSE (Persian Zero-Shot) - Factoid Reasoning Simple: leaderboard

Metric: Contains match (0-1; answer contains the gold string, split over any extra answers; 600 reasoning questions; 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 Instruct0.34
2Qwen 2.5 72B Instruct Turbo0.3
3Gemma 2 27B (IT)0.29
4Mistral Small 30.29

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