SEA-LION V1 3B

Training compute

2.2×10²² FLOP

Parameters

3B

Published

Dec 1, 2023

SEA-LION V1 3B is an AI model developed by AI Singapore (Singapore), first published in December 2023. It works in the language domain, on tasks such as question answering, chat and language modeling/generation.

Training it took an estimated 2.2×10²² FLOP of compute (estimation method: hardware,operation counting). The model has 3,000,000,000 parameters. It was trained on roughly 980B datapoints. Training ran on 240 NVIDIA A100 SXM4 40 GB for about 336 hours.

Access: Open weights (unrestricted). Its weights are openly available. Epoch AI rates the confidence of this record as confident.

Full record

Organization: AI Singapore
Country of organization: Singapore
Domain: Language
Task: Question answering, Chat, Language modeling/generation
Training compute: 2.2×10²² FLOP
Compute estimation method: Hardware, Operation counting
Parameters: 3,000,000,000
Dataset size: 980B
Training hardware: NVIDIA A100 SXM4 40 GB
Chips used: 240
Training time: 336 h
Training power draw: 190.4 kW
Numerical format: BF16
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Epoch confidence: Confident