Thai LLM NLG - flores200_eng_Latn_tha_Thai_seacrowd_t2t (BLEU): leaderboard
The Thai LLM Leaderboard, run as an open results dataset by the Thai LLM community. Generation quality on FLORES-200 translation from English into Thai, scored by BLEU (SEACrowd harness).
Metric: BLEU. Source: huggingface.co. Saturation forecast: Estimated already saturated. 68 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o (2024-05-13) | 34.18 | #325 |
| 2 | Claude 3.5 Sonnet (20240620) | 33.98 | #299 |
| 3 | GPT-4o Mini (2024-07-18) | 30.44 | #558 |
| 4 | Llama 3.1 70B Cpt Sea Lionv3 Instruct | 28.92 | #548 |
| 5 | Llama 3.1 70B Instruct | 28.08 | #561 |
| 6 | Llama 3.1 Nemotron 70B Instruct | 27.42 | #554 |
| 7 | Llama 3.3 70B Instruct | 27.26 | #534 |
| 8 | Llama 3 8B Cpt Sea Lionv2 Instruct | 27.25 | #1188 |
| 9 | gemma2-9B-cpt-sea-lionv3-instruct | 26.73 | #705 |
| 10 | DeepSeek R1 Distill Llama 70B | 25.53 | #594 |
| 11 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 25.3 | #1024 |
| 12 | Llama 3.1 8B CPT SEA LIONv3 Instruct | 24.95 | #810 |
| 13 | Gemma 2 9B (IT) | 24.72 | #788 |
| 14 | Qwen 2.5 72B Instruct | 24.6 | #454 |
| 15 | SeaLLMs-v3-7B-Chat | 23.37 | #915 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=thai-llm-nlg-flores200-eng-latn-tha-thai-seacrowd-t2t-bleu · How It Works · Data refreshed daily, snapshot 2026-10-20.