AI Models / Compare

Phi-4 Mini

Tiny but capable 3.8B Phi model for on-device and low-latency inference.

Creator: Microsoft
Lifecycle: Active
Context: 128.0K
Max output: 16.4K
Released: Mar 1, 2025
Status: unknown

Input: $0.04 / 1M tokens
Output: $0.13 / 1M tokens
Cached read: — / 1M tokens
Cached write: — / 1M tokens
Batch discount: —%
Source: Phi-4 Mini pricing
Verified: Apr 5, 2026 (High)

Capabilities

Modalities: text→text
Capabilities: batchSupportfunctionCallingstructuredOutputs
Strengths: 128K context in 3.8B, On-device friendly
Tradeoffs: Lower reasoning ceiling than Phi-4

Official Links

Intro

https://azure.microsoft.com/en-us/blog/phi-4-mini-small-but-mighty-reasoning-model/

Docs

https://azure.microsoft.com/en-us/products/phi

Pricing

https://techcommunity.microsoft.com/blog/azure-ai-foundry-blog/announcing-new-phi-pricing-empowering-your-business-with-small-language-models/4395112

Release note

https://azure.microsoft.com/en-us/updates/?query=phi

Benchmark Coverage

Benchmark	Version	Score	Date	Source	Notes

Release History

Release	Alias	Lifecycle	Release Date	Deprecation	Shutdown	Summary
Phi-4 Mini	phi-4-mini	Active	Mar 1, 2025	—	—	Current published model family snapshot.

Host Coverage

Host	Type	Context	Pricing Note	Differences
Azure AI Foundry	first-party	128.0K	$0.04/$0.13 per MTok.	—

Migration Guidance

Best tiny model for on-device. Upgrade to Phi-4 for more demanding tasks.

Replacement models: phi-4

Change Events

Date	Type	Title	Description	Source
Mar 1, 2025	family_added	Phi-4 Mini published	Initial public model family launch.	Phi-4 Mini release notes

Other models from Microsoft