DITING - Tense Consistency: leaderboard
DITING subtask scoring whether translations maintain consistent and correct tense across narrative passages.
Metric: Score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3 | 5.46 | #321 |
| 2 | GPT-4o | 5.22 | #343 |
| 3 | Qwen 3 32B | 5.1 | #443 |
| 4 | Qwen 3 14B | 4.9 | #537 |
| 5 | Qwen 3 8B | 4.66 | #674 |
| 6 | Seed-X-PPO-7B | 4.64 | #575 |
| 7 | Llama 3 70B | 4.34 | #631 |
| 8 | Llama 3 8B | 3.98 | #1242 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=diting-tense-consistency · How It Works · Data refreshed daily, snapshot 2026-10-20.