Swallow

Parameters

70M

Published

Apr 27, 2024

Swallow is an AI model developed by Tokyo Institute of Technology (Japan), first published in April 2024. It works in the language domain, on tasks such as language modeling/generation, chat and translation.

Epoch AI has no training-compute estimate for this model. The model has 70,000,000 parameters. It was trained on roughly 100B datapoints. Training ran on NVIDIA A100.

Access: Open weights (restricted use). Its weights are openly available. It is built on top of Llama 2-70B. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Tokyo Institute of Technology
Country of organization: Japan
Domain: Language
Task: Language modeling/generation, Chat, Translation
Compute estimation method: Reported, Operation counting
Parameters: 70,000,000
Dataset size: 100B
Training hardware: NVIDIA A100
Model accessibility: Open weights (restricted use)
Open weights: Yes
Base model: Llama 2-70B
Epoch confidence: Confident