🇺🇸MicrosoftOpen source

Microsoft: Phi 4.

See how Microsoft: Phi 4 compares with other broadly useful AI models across quality, speed, price, context, and practical capabilities.

CompareVisit model

Quality

#50/67

0

Overall quality score

Speed

#38/67

65 t/s

Output tokens per second

Input price

#11/67

$0.07/1M

Per 1M tokens

Output price

#6/67

$0.14/1M

Per 1M tokens

Context

#66/67

16.4K

Content read at once

Comparison summary

Microsoft: Phi 4 ranks #50 of 67 for overall quality and #38 for response speed in the current catalogue.

Its listed input price is $0.07/1M, output price is $0.14/1M, and it can work with up to 16.4K of context at once.

Its strongest areas in the current data are everyday help, coding, complex tasks.

Practical specifications

Company
Microsoft
Reasoning
No
Open weights
Yes
Input
Text
Output
Text
Context window
16.4K

Microsoft: Phi 4 technical benchmarks.

Named evaluation records for technical comparison. Results retain their original benchmark names and sources.

BenchmarkAreaPhi 4Source
GPQAGPQA Diamond57.5OpenRouter model benchmarks
Humanity's Last ExamHLE4.1OpenRouter model benchmarks
IFBenchIFBench23.5OpenRouter model benchmarks
τ²-Bench Telecomτ²-Bench Telecom0OpenRouter model benchmarks
AA-LCRAA-LCR0OpenRouter model benchmarks
CritPtCritPt0OpenRouter model benchmarks
SciCodeSciCode26OpenRouter model benchmarks
Terminal-Bench HardTerminal-Bench Hard3.8OpenRouter model benchmarks
AA-Omniscience AccuracyAA-Omniscience Accuracy13.2OpenRouter model benchmarks
AA-Omniscience Non-Hallucination RateAA-Omniscience Non-Hallucination Rate19.5OpenRouter model benchmarks

Quality benchmarks

See where Microsoft: Phi 4 stands.

The darker bar marks Microsoft: Phi 4; the lighter bars provide context from other leading models.

Highlights

Quality

Overall capability · Higher is better

Speed

Output tokens per second · Higher is better

Input price

USD per 1M tokens · Lower is better