CanMT - Spanish to English: leaderboard
Metric: Translation score (1-7) on the 134 Spanish-to-English 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 March 2027. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 | 5.24 |
| 2 | Grok 4.1 | 5.22 |
| 3 | DeepSeek R1 | 5.16 |
| 4 | DeepSeek V3.2 | 5.1 |
| 5 | GPT-4o | 5.07 |
| 6 | Gemini 2.5 Flash Lite | 5.05 |
| 7 | Llama 3.3 70B Instruct | 5.04 |
| 8 | Qwen 2.5 72B Instruct | 5.01 |
| 9 | Qwen 2.5 32B Instruct | 4.99 |
| 10 | Qwen 3 32B (Non-reasoning) | 4.92 |
| 11 | Qwen 2.5 14B Instruct | 4.89 |
| 12 | Qwen 3 14B (Non-reasoning) | 4.85 |
| 13 | Mixtral 8x7B Instruct (v0.1) | 4.81 |
| 14 | Qwen 2.5 7B Instruct | 4.61 |
| 15 | Qwen 3 8B (Non-reasoning) | 4.5 |
Interactive version: theaggregate.ai/benchmark?slug=canmt-spanish-to-english · How It Works · Data refreshed daily, snapshot 2026-10-07.