CIBuzzBench - Equivalent Selection (Chinese Prompt): leaderboard

Metric: Macro F1 (%; four-way choice of the English equivalent against literal, cultural-mismatch and pragmatic-neighbor distractors, mean over five option-shuffle seeds; Chinese instructions; 3,001 Chinese internet buzzwords; zero-shot, temperature 0; the Chinese buzzword is the input under both prompt languages). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)88.41
2GPT-5.587.83
3Qwen 3.7 Max84.8
4GLM-5.284.4
5Claude Opus 4.883.17
6DeepSeek V4 Pro78.91
7Qwen 3 14B56.98
8Qwen 3 8B54.91
9Qwen 3 4B45.41

Interactive version: theaggregate.ai/benchmark?slug=cibuzzbench-equivalent-selection-chinese-prompt · How It Works · Data refreshed daily, snapshot 2026-09-26.