Komodo-7B
Indonesian LLM covering English and 11 regional languages.
About
Komodo-7B is a 7 billion parameter language model from Yellow.ai's NLP research team, introduced with an arXiv paper in March 2024. It was built by incremental pretraining and vocabulary expansion on top of Llama-2-7B, trained on over 8.5 billion Indonesian and regional-language tokens covering Indonesian, English, and 11 regional languages including Javanese, Sundanese, Balinese, Minangkabau, and Acehnese. Base and instruct variants were released.
The team reports that Komodo-7B-Instruct outperformed GPT-3.5, Aya-101, Llama-2-Chat-13B, and Mixtral-8x7B-Instruct on their Indonesian evaluation suite. License caveat: the weights carry the Llama 2 community license terms rather than a fully open license, so commercial use is restricted by Meta's license conditions.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 115; Trending score 68/100 from researcher assessment.