Microsoft: Phi-4-mini

Compact Microsoft instruction model tuned for efficient coding assistance, reasoning, and low-latency agent tasks

Model lab: Microsoft
Model family: phi

Model resources

Weights

Modalitiestext → text

Context window128,000

Maximum output4,096

InputUSD 0.06 / 1M tokens

OutputUSD 0.24 / 1M tokens

Try now

Quickstart

Use your Moduck API Key, set the API base URL to https://api.moduck.ai/v1, and set the model parameter to microsoft/phi-4-mini.

import os
import requests

base_url = "https://api.moduck.ai/v1"
api_key = os.environ["MODUCK_API_KEY"]
model_id = "microsoft/phi-4-mini"

response = requests.post(
    f"{base_url}/chat/completions",
    headers={
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json",
    },
    json={
        "model": model_id,
        "messages": [{"role": "user", "content": "Hello!"}],
        "stream": False,
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

Use the familiar OpenAI-style chat request format to make model integration easier.

Phi-4-mini API Pricing

Type	Price information
Input	USD 0.06 / 1M tokens
Output	USD 0.24 / 1M tokens

Benchmarks

Benchmark	Score	Measurement	Source
MMLU	67.3	accuracy

Performance

Performance trends

Throughput90.8 tok/sCurrent

Time to first token1.3 sCurrent

End-to-end latency6.9 sCurrent

Throughput

Current90.8 tok/s

Window average89.4 tok/s

Time to first token

Current1.3 s

Window average1.5 s

End-to-end latency

Current6.9 s

Window average7.5 s

Tail latency (P99)

Current14.6 s

Window average16.6 s

Prompt cache hit rate

Current19.1%

Window average23.0%

Error rate

Current0.0%

Window average0.1%

Availability

Uptime100%At least one provider responded

Availability100%Requests actually served

Error rate0%Share of failed requests

Served requests, hour by hour

24 hours agoNow

Phi-4-mini Compare models

Model	Input price	Output price	Context window	Capabilities
Phi-4-miniThis page	USD 0.06 / 1M tokens	USD 0.24 / 1M tokens	128,000	Text output · Tool calling
MAI-Code-1-Flash	USD 0.6 / 1M tokens	USD 3.6 / 1M tokens	256,000	Text output · Reasoning · Tool calling
WizardLM 2 8x22B	USD 0.384 / 1M tokens	USD 0.384 / 1M tokens	65,536	Text output
Beam 1.0 501B A23B	USD 0.2 / 1M tokens	USD 0.8 / 1M tokens	1,048,576	Text output · Reasoning · Tool calling
Ling 3.1 Flash	USD 0.06 / 1M tokens	USD 0.176 / 1M tokens	262,144	Text output · Reasoning · Tool calling

Model specifications

Input: text
Output: text
Context window: 128,000
Maximum output: 4,096
Release date: 2024-12-11
Attachments: No
Reasoning: No
Tool calling: Yes
Temperature setting: Yes
Open weights: Yes

Frequently asked questions

What is Phi-4-mini for?

Compact Microsoft instruction model tuned for efficient coding assistance, reasoning, and low-latency agent tasks

What inputs and outputs does Phi-4-mini support?

Recorded inputs: text; outputs: text.

How is Phi-4-mini priced?

The pricing table shows the current recorded Moduck price. For example: Input USD 0.06 / 1M tokens; Output USD 0.24 / 1M tokens. See the pricing table for all units, variants, and conditions.

What specifications are recorded for Phi-4-mini?

Current model records list Context window 128,000 tokens; Maximum output 4,096 tokens.

What model parameter does the Quickstart use?

The example uses microsoft/phi-4-mini. This value comes from the page identifier and is not confirmed as a callable Moduck public ID.

Can I call this model through the Moduck API now?

This page has no verified Moduck API call for this model. A Quickstart example or reserved ID does not establish access; check the models available to your account.