CanMT - Chinese to Spanish: leaderboard
CanMT (2026) is a culture-aware machine translation benchmark of novel sentences containing culture-specific items, scored by an LLM judge with references on five translation-quality dimensions. This board is the translation score on a 1 to 7 scale for Chinese to Spanish; higher is better.
Metric: Translation score (1-7) on the 146 Chinese-to-Spanish CanMT sentences containing culture-specific items, GPT-5-nano judge with the reference translation scoring contextual accuracy, cultural adaptation, functional equivalence, fidelity and naturalness on 7-point Likert scales, averaged per sentence; default translation prompt; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 22 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Grok 4.1 | 5.25 | #462 | |
| 2 | GPT-4 | 5.21 | #970 | |
| 3 | Gemini 2.5 Flash Lite | 5.16 | #964 | |
| 4 | DeepSeek V3.2 | 5.1 | #419 | |
| 5 | Seed-X-PPO-7B | 5.07 | ||
| 6 | DeepSeek R1 | 5.05 | #482 | |
| 7 | Qwen 2.5 72B Instruct | 5.02 | #1117 | |
| 8 | GPT-4o | 4.92 | #751 | |
| 9 | Llama 3.3 70B Instruct | 4.8 | #1385 | |
| 10 | Qwen 3 14B | Non-reasoning | 4.78 | #1269 |
| 11 | Qwen 2.5 32B Instruct | 4.74 | #1247 | |
| 12 | Qwen 3 32B | Non-reasoning | 4.71 | #972 |
| 13 | Google Translate | 4.61 | ||
| 14 | Qwen 2.5 14B Instruct | 4.53 | #1775 | |
| 15 | Qwen 3 8B | Non-reasoning | 4.38 | #1873 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=canmt-chinese-to-spanish · How It Works · Data refreshed daily, snapshot 2026-11-21.