Thai LLM NLG - flores200_tha_Thai_eng_Latn_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 Thai into English, 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 | Claude 3.5 Sonnet (20240620) | 54.81 | #308 |
| 2 | GPT-4o (2024-05-13) | 52.96 | #337 |
| 3 | Llama 3.1 70B Instruct | 52.76 | #574 |
| 4 | Llama 3.1 70B Cpt Sea Lionv3 Instruct | 52.38 | #559 |
| 5 | Qwen 2.5 72B Instruct | 52.01 | #470 |
| 6 | Llama 3 8B Cpt Sea Lionv2 Instruct | 50.77 | #1174 |
| 7 | GPT-4o Mini (2024-07-18) | 50.45 | #567 |
| 8 | DeepSeek R1 Distill Llama 70B | 49.94 | #615 |
| 9 | Gemma 2 9B (IT) | 49.88 | #794 |
| 10 | Llama 3.3 70B Instruct | 49.75 | #541 |
| 11 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 49.59 | #1011 |
| 12 | Qwen 2.5 32B Instruct | 49.55 | #511 |
| 13 | Llama 3.1 8B Instruct | 49.53 | #1019 |
| 14 | Llama 3.1 8B CPT SEA LIONv3 Instruct | 49.3 | #817 |
| 15 | DeepSeek R1 Distill Qwen 32B | 48.34 | #662 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=thai-llm-nlg-flores200-tha-thai-eng-latn-seacrowd-t2t-bleu · How It Works · Data refreshed daily, snapshot 2026-11-06.