Model Information

Display Name: Phi-4

API Model ID: microsoft/phi-4

Category: Text To Text

Description: Microsoft's Phi-4 is a 14B model engineered for reasoning. Trained on carefully curated and synthetic data, it rivals far larger models on math, logic, and STEM tasks while staying compact and affordable. **Key Features:** - 16K token context window - Exceptional math and step-by-step reasoning for its size - Strong on STEM, logic, and structured problems - JSON mode for structured outputs - Streaming support **Best For:** - Math and logical reasoning at low cost - STEM and science Q&A - Code logic and structured problem solving - Compact reasoning pipelines

Context Window: 16,384 tokens

Max Output: 4,096 tokens

How to Use This Model

To use Phi-4 via the HInow.ai API, use the model ID: microsoft/phi-4

API Request Example (Chat/Text)

POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "microsoft/phi-4",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}
              

Pricing

• input: $0.65
• output: $0.65

Available Parameters

• temperature: Controls randomness (0-2). Default: 0.7 (Options: 0, 0.3, 0.5, 0.7, 1.0, 1.5, 2.0)
• top_p: Nucleus sampling (0-1). Default: 0.9 (Options: 0.1, 0.5, 0.7, 0.9, 0.95, 1.0)
• max_tokens: Max tokens to generate (1-4096) (Options: 256, 512, 1024, 2048, 4096)
• response_format: Output format (Options: text, json_object)

Quick Reference

To use this model, set: "model": "microsoft/phi-4"

Featured: No

Documentation: https://hinow.ai/models/microsoft/phi-4

API Endpoint: https://api.hinow.ai/v1