Konan LLM 41B

Training compute

1.7×10²³ FLOP

Parameters

41B

Published

Dec 15, 2023

Konan LLM 41B is an AI model developed by Konan Technology (South Korea), first published in December 2023. It works in the language and vision domain, on tasks such as language modeling/generation.

Training it took an estimated 1.7×10²³ FLOP of compute (estimation method: operation counting). The model has 41,000,000,000 parameters. It was trained on roughly 7T datapoints.

Access: Hosted access (no API). Its weights are not openly released. Epoch AI rates the confidence of this record as likely.

Full record

Organization: Konan Technology
Country of organization: South Korea
Domain: Language, Vision
Task: Language modeling/generation
Training compute: 1.7×10²³ FLOP
Compute estimation method: Operation counting
Parameters: 41,000,000,000
Dataset size: 7T
Model accessibility: Hosted access (no API)
Open weights: No
Epoch confidence: Likely