Skip to content
NINJA

Phi 4

Released by Microsoft on 10 Jan 2025.

API id microsoft/phi-4

Input price
$0.07 / 1M tokens
Output price
$0.14 / 1M tokens
Context window
16K tokens
Longest reply
15K tokens

About this model

Microsoft Research 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...

Description from the model’s public catalogue listing.

Capabilities

Reads
Text
Writes
Text
Reasoning mode
No
Tool calling
No
Open weights
Yes
Cached input
Not listed

Electricity per answer

GPU energy measured by AI Energy Score. Compare with other models

Electricity, chat answer
0.029 Wh

What it would cost

At list price, without caching or batch discounts.

1,000 chat replies2K tokens in, 500 out each
$0.21
Summarise 100 long reports50K tokens in, 1K out each
$0.36
A coding agent’s working day20M tokens in, 1M out
$1.54
Price your own workload on every model

Price history

No change since we started tracking prices on 30 Sept 2026.

Alternatives at a similar price

From other labs, priced within about 40% of Phi 4, with a comparable context window.