Models/phi-4

phi-4Active · Cloud

DeepInfra · Released Jan 2025 · MIT

[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...

Context
16K
Max output
16K
Input
$0.07
Output
$0.14
Cached in
-
Latency
-

Capabilities

Core
Streaming
Function calling
JSON mode
Structured outputs
Multimodal
Vision
Image input
Audio
Image generation
Video
Advanced
Reasoning
Tool calling
MCP
Prompt caching
Batch API
Embeddings

Benchmarks

BBH52.4
GPQA20.8
IFEval5.9
MATH Lvl 531.6
MMLU-Pro47.6
MUSR23.8
Open LLM Average30.4

Details

ProviderDeepInfra
ReleasedJan 2025
LicenseMIT
InputText
OutputText
Throughput-
Availability100%

Related models