LLM Stats (WMT24++): leaderboard
Machine translation evaluation from WMT 2024 shared task covering 11 language pairs with human quality judgments.
Metric: Score (%). Source: llm-stats.com. Source: qwen.ai. Saturation forecast: Around January 2027. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Nemotron 3 Super | 86.67 | #335 |
| 2 | nemotron-3-nano-30B-a3B | 86.2 | #662 |
| 3 | Qwen 3.7 Max | 85.8 | #75 |
| 4 | Qwen 3.7 Plus | 84.6 | #92 |
| 5 | Qwen 3.6 Plus | 84.3 | #120 |
| 6 | Nemotron 3 Ultra | 83.7 | #235 |
| 7 | Claude Opus 4.6 (Thinking) | 82.7 | #63 (Claude Opus 4.6) |
| 8 | DeepSeek V4 Pro (Reasoning, Max Effort) | 82.2 | #102 (DeepSeek V4 Pro) |
| 9 | GLM-5.1 (Thinking) | 81.8 | #97 (GLM-5.1) |
| 10 | Kimi K2.6 (Thinking) | 81.6 | #104 (Kimi K2.6) |
| 11 | Command A+ | 81 | #413 |
| 12 | Qwen 3.5 397B A17B | 78.9 | #146 |
| 13 | Qwen 3.5 122B A10B | 78.3 | #173 |
| 14 | Qwen 3.5 27B | 77.6 | #187 |
| 15 | Qwen 3 Max (Thinking) | 77.6 | #208 (Qwen 3 Max) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-wmt24-plus-plus · How It Works · Data refreshed daily, snapshot 2026-10-20.