/ Models

Language models

Microsoft: Phi 4

[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...

Maker
Microsoft
Context window
16K
Max output
16K
Input / 1M
$0.07
Output / 1M
$0.14
1M in + 1M out
$0.21

Everything else

Max output16K
Input / 1M$0.07
Output / 1M$0.14
Cache read / 1M
Cache write / 1M
1M in + 1M out$0.21
Takestext
Returnstext
Toolsno
Reasoningno
Open weightsyes
Released2025-01-10
Knowledge cutoff2024-06-30
Retires
Catalog idmicrosoft/phi-4

Benchmarks

sourceepoch
eci131.07
confidenceConfident
accessibilityOpen weights (unrestricted)
gpqa0.561
gpqaStderr0.0259
mathLevel50.649
mathLevel5Stderr0.0107
aime0.138
aimeStderr0.037

Benchmark scores by Epoch AI, CC BY 4.0.

Using it

In 00, this model is picked per agent — and per tier, so the model that answers your customers need not be the one that writes your code. Get 00.

← all models