AI Hub
All models
LLMOpen weights

Phi-3

Microsoft

Updated May 21, 2026 · Benchmarks via Artificial Analysis, specs & pricing via OpenRouter · Methodology · Report an error

Specifications

Type
LLM
Access
Open weights
Released
April 23, 2024
License
MIT
Parameters
3.8B
Context window
128,000 tokens
Input
Text
Output
Text

Our take

A 3.8B model good enough to run on a phone, continuing the small-but-mighty, data-quality-first line of research.

Related research

Compare Phi-3 with

See all Phi-3 alternatives →