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...
Azure AI Foundry modelsModel Snapshot
ReleaseDec 12, 2024
Context16,384
Overall
Rank #513 across 600
4.9
Coding
No rank
n/a
Blended Price
Rank #63 across 379
$0.219/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed41.9 tok/s
First Token0.47s
Max Output16,384
Pricing
Blended$0.219/M
Input$0.125/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 4.9 | #513 |
| Artificial Analysis Math Index | math | 18.0 | #214 |
| MMLU-Pro | reasoning | 71.4% | #185 |
| GPQA | reasoning | 57.5% | #365 |
| Humanity's Last Exam |