HardMTBench - English to Chinese: leaderboard
HardMTBench (2026) is a difficulty-aware Chinese-English translation test of 10,000 pairs from 12 knowledge-intensive domains, selected for domain-knowledge density and translation difficulty. This board scores the English-to-Chinese direction with GEMBA-DA, an LLM direct assessment from 0 to 100; higher is better.
Metric: GEMBA-DA direct assessment score (0-100) of the English-to-Chinese translations, judged by gpt-oss-120b, on HardMTBench, 10,000 domain-balanced, difficulty-selected Chinese-English pairs from 12 knowledge-intensive domains, each translated in both directions under one unified protocol; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 16 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 92.78 | #99 | |
| 2 | GPT-5.5 | 92.66 | #54 | |
| 3 | Qwen 3.5 397B A17B | 92.13 | #264 | |
| 4 | DeepSeek V4 Pro | Thinking | 91.79 | #196 |
| 5 | Qwen 3.6 35B A3B | 91.05 | #481 | |
| 6 | Qwen-MT-Plus | 90.21 | ||
| 7 | Hy-MT2-30B-A3B | 89.71 | ||
| 8 | Gemma 4 31B (IT) | Thinking | 89.03 | #350 |
| 9 | gemma-4-26B-A4B-it | Thinking | 88.53 | |
| 10 | Gemma 4 26B A4B (IT) | 87.57 | #556 | |
| 11 | Hy-MT2-7B | 86.91 | ||
| 12 | Gemma 4 E4B (IT) | Thinking | 83.24 | |
| 13 | gemma-4-E4B-it | 82.82 | #1539 | |
| 14 | Hy-MT2-1.8B | 82.28 | ||
| 15 | Gemma 4 E2B (IT) | Thinking | 80.87 | #2451 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=hardmtbench-english-to-chinese · How It Works · Data refreshed daily, snapshot 2026-11-21.