CultureTalk-ID - Translation into Local Languages: leaderboard

Metric: BLEU-4 (0-100) of zero-shot dialogue translation from Indonesian into the local languages with province and language context, mean of three runs for open-weight models; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 9 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash28.36
2GPT-5.127.53
3gemma2-9B-cpt-sahabatai-v1-instruct21.7
4Command A21.45
5Gemma 2 9B (IT)20.54
6Qwen 3 8B19.78
7Llama 3.1 8B Instruct17.62
8Sailor2-8B-Chat15.44

Interactive version: theaggregate.ai/benchmark?slug=culturetalk-id-translation-into-local-languages · How It Works · Data refreshed daily, snapshot 2026-09-29.