Phi-4-multimodal-instruct

MicrosoftMCP Tool Call

microsoft/phi-4-multimodal-instructOpen Weights

Compact Microsoft multimodal model for text, image, and speech understanding

Benchmark Pricing$0.07 / 1M in$0.11 / 1M out$0.04 cache read

Context: 131,072 tokens•Max Output: 131,072 tokens•Knowledge: 2024-06•Released: 2025-02-26

TECHNICAL SPECIFICATIONS

Context Window131,072131.1K tokens

Max Output131,072131.1K tokens

LicenseMITSelf-hostable

Input Modalities

TextVisionAudio

Function / Tool CallingSupported (MCP Ready)

Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Phi-4-multimodal-instruct natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 131,072 tokens, this model can handle moderately large MCP tool definitions and multi-turn conversation logs comfortably.

API CALL INVOCATION EXAMPLE

// Example: Calling Phi-4-multimodal-instruct via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.MICROSOFT_API_KEY,
});

const response = await client.chat.completions.create({
  model: "microsoft/phi-4-multimodal-instruct",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);