modelgrep
M

Microsoft: Phi 4 Mini Instruct

microsoft/phi-4-mini-instruct

172nd smartest of 178Cheaper than 88% of paidJSON
Use via OpenRouter ↗
Intelligence
8.4
172nd of 178
Design Elo
Speed
199
16th fastest
Latency
166ms
first token
Input price
$0.080
37th cheapest
Context
131K
128K max out

How it compares

Smarter than3%
of all ranked models
Faster than95%
of all ranked models
Cheaper than88%
of all ranked models

Overview

Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data. The model belongs to the Phi-4...

Benchmarks

independent · via OpenRouter
Artificial Analysis6th percentile
Intelligence Index
8.4
Coding Index
3.6
Agentic Index
2.7
GPQA Diamond
33%
Humanity's Last Exam
4%
SciCode
11%
Tau²-Bench (agentic)
8%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
WandBbf16$0.080$0.350

Specifications

Context window131K
Max output128K
Knowledge cutoff
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price$0.080/M
ModeratedNo

Phi 4 Mini Instruct FAQ

How much does Phi 4 Mini Instruct cost?

Phi 4 Mini Instruct costs $0.080 per million input tokens and $0.350 per million output tokens via OpenRouter, making it 37th cheapest of 298 paid models.

How smart is Phi 4 Mini Instruct?

Phi 4 Mini Instruct scores 8.4 on the Artificial Analysis Intelligence Index, ranking 172nd of 178 benchmarked models, with a GPQA Diamond score of 33%.

How fast is Phi 4 Mini Instruct?

Phi 4 Mini Instruct generates around 199 tokens per second with 166ms time-to-first-token (p50), the 16th fastest tracked model.

What is Phi 4 Mini Instruct's context window?

Phi 4 Mini Instruct supports a 131K-token context window and can output up to 128K tokens. It accepts text input.

Compare head-to-head