Phi-4-Reasoning

Training compute

9.3×10²³ FLOP

Parameters

14B

Published

Apr 30, 2025

Phi-4-Reasoning is an AI model developed by Microsoft (United States), first published in April 2025. It works in the language domain, on tasks such as language modeling/generation, quantitative reasoning, question answering and 2 more.

Training it took an estimated 9.3×10²³ FLOP of compute (estimation method: operation counting,hardware). The model has 14,000,000,000 parameters. Training ran on 32 NVIDIA H100 SXM5 80GB for about 60 hours.

Access: Open weights (unrestricted). Its weights are openly available. It is built on top of Phi-4. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Microsoft
Country of organization: United States
Domain: Language
Task: Language modeling/generation, Quantitative reasoning, Question answering, Mathematical reasoning, Code generation
Training compute: 9.3×10²³ FLOP
Compute estimation method: Operation counting, Hardware
Parameters: 14,000,000,000
Training hardware: NVIDIA H100 SXM5 80GB
Chips used: 32
Training time: 60 h
Training power draw: 43.9 kW
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Base model: Phi-4
Epoch confidence: Confident