Skip to main content
Microsoft

Phi-4 Mini Instruct

Phi-4 Mini Instruct by Microsoft — compare inference API pricing across providers.

Contact providers for pricing

API Pricing

No pricing data available for this model at the moment.

Prices updated daily. Last check: Aug 28, 2026

Performance & Benchmarks

Source: Artificial Analysis →
Intelligence
5.7 / 100
Coding
3.8 / 100
Math
6.7 / 100
Output Speed
45.7 t/s
Latency (TTFT)
388ms

Reasoning & Knowledge

  • MMLU-Pro46.5%
  • GPQA Diamond33.1%
  • Humanity's Last Exam4.4%

Coding

  • LiveCodeBench12.6%
  • SciCode10.8%

Math

  • AIME 20256.7%
  • AIME3.0%
  • MATH-50069.6%

Agentic & Tool Use

  • Terminal-Bench Hard0.0%
  • Terminal-Bench v2.10.4%
  • τ²-bench8.2%

Instruction & Long Context

  • IFBench21.1%
  • Long-Context Reasoning16.7%

Benchmarks measured Aug 2026. Scores are independent evaluations, not vendor-reported.

Model Details

General

Creator
Microsoft
Modalities
Text

Capabilities

Tool Calling
No
Open Source
No