CultureTalk-ID - Translation from Local Languages: leaderboard

Metric: BLEU-4 (0-100) of zero-shot dialogue translation from the local languages into Indonesian with province and language context, mean of three runs for open-weight models; higher is better. Source: arxiv.org. Saturation forecast: Around October 2027. 9 models tracked.

Top models

#ModelScore
1Gemini 3 Flash (Preview)52.74
2GPT-5.150.54
3Command A42
4gemma2-9B-cpt-sahabatai-v1-instruct39.2
5Gemma 2 9B (IT)30.41
6Llama 3.1 8B Instruct26.6
7Qwen 3 8B24.75
8Sailor2-8B-Chat22.84

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