phi-3-medium 14B

Training compute

4×10²³ FLOP

Parameters

14B

Published

Apr 23, 2024

phi-3-medium 14B is an AI model developed by Microsoft (United States), first published in April 2024. It works in the language domain, on tasks such as chat and language modeling/generation.

Training it took an estimated 4×10²³ FLOP of compute (estimation method: operation counting). The model has 14,000,000,000 parameters. It was trained on roughly 4.8T datapoints.

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

Full record

Organization: Microsoft
Country of organization: United States
Domain: Language
Task: Chat, Language modeling/generation
Training compute: 4×10²³ FLOP
Compute estimation method: Operation counting
Parameters: 14,000,000,000
Dataset size: 4.8T
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Epoch confidence: Likely

Benchmark results

01

Epoch Capabilities Index

ECI Score

121