# Phi 4 Reasoning API Pricing

As of October 10, 2026, Phi 4 Reasoning costs $0.125 per 1M input tokens and $0.50 per 1M output tokens on Azure AI Foundry. It has a 32,768-token context window and returns up to 4,096 tokens per response.

- Page: https://models.litellm.ai/models/phi-4-reasoning
- Lab: Microsoft
- LiteLLM model name: `azure_ai/Phi-4-reasoning`
- Context window: 32,768 tokens
- Max output: 4,096 tokens
- Added to LiteLLM: October 2025
- Prices updated: October 10, 2026

## Pricing by provider

Token prices in USD per 1M tokens.

| Provider | Model name | Input | Output | Cache read | Per image | Per second | Context |
|---|---|---|---|---|---|---|---|
| Azure AI Foundry | `azure_ai/Phi-4-reasoning` | $0.125 | $0.50 | – | – | – | 32K |

## Cost examples

1,000 requests with 2,000 input and 500 output tokens each:

- Azure AI Foundry: $0.50

## Features

Supported: Function calling, Reasoning.

## Use it with LiteLLM

```python
from litellm import completion

response = completion(
    model="azure_ai/Phi-4-reasoning",
    messages=[{"role": "user", "content": "Hello!"}],
)
```

## FAQ

### How much does Phi 4 Reasoning cost?

Phi 4 Reasoning costs $0.125 per 1M input tokens and $0.50 per 1M output tokens on Azure AI Foundry.

### What is the context window of Phi 4 Reasoning?

Phi 4 Reasoning has a 32,768-token context window and returns up to 4,096 tokens per response.

### Which providers offer Phi 4 Reasoning?

Phi 4 Reasoning is available from Azure AI Foundry through LiteLLM, using the model name azure_ai/Phi-4-reasoning.

### How do I call Phi 4 Reasoning with an OpenAI-compatible API?

Use the LiteLLM Python SDK with model="azure_ai/Phi-4-reasoning", or add the model to the LiteLLM proxy's config.yaml and send requests to /v1/chat/completions from any OpenAI client. LiteLLM tracks the cost of every request at the prices on this page.

### What features does Phi 4 Reasoning support?

Phi 4 Reasoning supports function calling and reasoning.

### When did LiteLLM add Phi 4 Reasoning?

Phi 4 Reasoning was added to LiteLLM's model price file in October 2025.

Source: LiteLLM model price file, https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json