G-IdiomAlign - Multiple-Choice Equivalence - Chinese 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 Chinese). Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V3.2 (Thinking) | 69.37 |
| 2 | Gemini 2.5 Pro | 67.77 |
| 3 | DeepSeek V3.2 (Non-reasoning) | 56.84 |
| 4 | Claude Haiku 4.5 | 56.8 |
| 5 | Qwen 3 8B (Thinking) | 41.2 |
| 6 | Qwen 3 8B (Non-reasoning) | 32.69 |
| 7 | Ministral-8B-Instruct-2410 | 28.63 |
Interactive version: theaggregate.ai/benchmark?slug=g-idiomalign-multiple-choice-equivalence-chinese-targets · How It Works · Data refreshed daily, snapshot 2026-09-26.