Requesty
DeepInfra Inc. logo
DeepInfra Inc.

phi-4

Phi-4-reasoning-plus is an enhanced 14B parameter model from Microsoft, fine-tuned from Phi-4 with additional reinforcement learning to boost accuracy on math, science, and code reasoning tasks. It uses the same dense decoder-only transformer architecture as Phi-4, but generates longer, more comprehensive outputs structured into a step-by-step reasoning trace and final answer. While it offers improved benchmark scores over Phi-4-reasoning across tasks like AIME, OmniMath, and HumanEvalPlus, its responses are typically ~50% longer, resulting in higher latency. Designed for English-only applications, it is well-suited for structured reasoning workflows where output quality takes priority over response speed.

Try phi-4CompareDocs

Specifications

Context window16K tokens
Max output—
API typechat
AddedMay 1, 2025
Model IDdeepinfra/microsoft/phi-4
Data retentionNo
Used for trainingNo
Provider location🇺🇸 US
Privacy policyDeepInfra Privacy Policy →

Benchmarks

Benchmarks haven't been published yet for this exact variant.

Some variants (region-specific deployments, highspeed tiers) share benchmarks with their base model — check the base model page or the DeepInfra Inc. models overview.

Pricing

Input / 1M
$0.07
Output / 1M
$0.14
Cache write
—
Cache read
—
Estimated cost
100K input + 10K output$0.0084
1M input + 100K output$0.0840
10M input + 1M output$0.84

Requesty charges exactly what the upstream provider charges — no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to deepinfra/microsoft/phi-4.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="deepinfra/microsoft/phi-4",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other DeepInfra Inc. models

Kimi K2.5
262K ctx$0.45/1M
deepseek-ai/DeepSeek-V3.1
164K ctx$0.30/1M
zai-org/GLM-4.5-Air
131K ctx$0.20/1M
zai-org/GLM-4.5
131K ctx$0.60/1M
Qwen/Qwen3-Coder-480B-A35B-Instruct
262K ctx$0.40/1M
Qwen/Qwen2.5-72B-Instruct
131K ctx$0.23/1M

Frequently asked questions

How much does phi-4 cost?
phi-4 is priced at $0.07 per million input tokens and $0.14 per million output tokens when accessed via Requesty. Requesty charges exactly what the upstream provider charges — we don't add markup.
What is the context window of phi-4?
phi-4 has a context window of 16K tokens. That's roughly 22 words of input you can fit in a single prompt.
What can phi-4 do?
phi-4 is a text-generation model you can call through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use phi-4 with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "deepinfra/microsoft/phi-4". The Quickstart above shows Python, JavaScript and cURL snippets.

Access phi-4 through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll DeepInfra Inc. models