TL

Trillion Labs

1-50 employees· South Korea·Website ↗

Korean AI startup building sovereign large language models and data curation infrastructure for the Korean language.

Technology
Share:
Get the full context.

Sign up to read complete case studies, access detailed metrics, and unlock all use cases.

Use Cases (1)

TL
Trillion Labs
7x
data processing speedup

Trillion Labs, a Korean AI startup building sovereign LLMs for the Korean language, deployed NVIDIA NeMo Curator to accelerate data curation across 2 trillion tokens. GPU-accelerated processing on 8x H100s cut processing time from 24 hours to 3.4 hours — a 7x improvement — and reduced compute costs up to 10x compared to CPU pipelines, while delivering a 5% accuracy boost for Korean language models.

TechnologyNNNVIDIA NeMo CuratorNNNVIDIA NeMo
Trillion Labs AI Use Cases & Strategy | Applied