Phi-4-multimodal API
About
Integrate multimodal vision capabilities using Phi-4-multimodal, designed to process both text and images seamlessly. This model delivers cost-effective pricing at $0.08/1M input and $0.32/1M output tokens, open weights architecture. Access Phi-4-multimodal via the Azure Cognitive Services API with up to 4K output tokens.
Capabilities
Input Modalities
textimageaudio
Output Modalities
text
Reasoning
No
Structured Output
No
Tool Use
No
WeightsOpen
Temperature
Adjustable
Attachment
Supported
Limits
Context Window
128K
Tokens
Input Limit
128K
Tokens
Max Output
4K
Tokens
Pricing
Standard (per 1M tokens)
Input
$0.08
Output
$0.32
Audio (per 1M tokens)
Input
$4.00
Frequently Asked Questions
Phi-4-multimodal by Azure Cognitive Services costs $0.08 per 1M input tokens and $0.32 per 1M output tokens.
Model Details
ID
phi-4-multimodal
Provider
Azure Cognitive ServicesFamily
phi
Release Date
Dec 11, 2024
Knowledge Cutoff
Oct 1, 2023
API Integration
NPM Package
@ai-sdk/azure
Environment Variables
AZURE_COGNITIVE_SERVICES_RESOURCE_NAME
AZURE_COGNITIVE_SERVICES_API_KEY