G-IdiomAlign - Multiple-Choice Equivalence - English Targets: leaderboard

Metric: Accuracy (%; source idiom to its meaning-equivalent target-language idiom among four options, the other three typed distractors (literal translation, lexical cue, contextual association), greedy decoding; macro-average over the 8 directions into English). Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.

Top models

#ModelScore
1DeepSeek V3.2 (Thinking)66.7
2Gemini 2.5 Pro66.53
3DeepSeek V3.2 (Non-reasoning)57.84
4Claude Haiku 4.555.54
5Qwen 3 8B (Thinking)39.57
6Qwen 3 8B (Non-reasoning)33.95
7Ministral-8B-Instruct-241030.29

Interactive version: theaggregate.ai/benchmark?slug=g-idiomalign-multiple-choice-equivalence-english-targets · How It Works · Data refreshed daily, snapshot 2026-09-26.