PaliBench - BLEU: leaderboard
Metric: BLEU (0-100), sacrebleu corpus BLEU, 1,700 Pali canon passages translated into English, each scored against three human reference translations (Bodhi, Sujato, Thanissaro), corpus level, models through OpenRouter; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 63.3 |
| 2 | Gemini 3 Flash (Preview) | 57.9 |
| 3 | Claude Opus 4.5 | 56.2 |
| 4 | DeepSeek V3.2 | 51.3 |
| 5 | Kimi K2.5 | 47.5 |
| 6 | GLM-4.7 | 43.7 |
| 7 | GPT-5.2 | 42.5 |
| 8 | Qwen 3 235B A22B 2507 Instruct | 40.6 |
| 9 | Grok 4.1 Fast | 40.1 |
| 10 | Llama 3.3 70B Instruct | 34.4 |
Interactive version: theaggregate.ai/benchmark?slug=palibench-bleu · How It Works · Data refreshed daily, snapshot 2026-10-07.