Claude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/MClaude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/M
BETA
Subscribe
← Back to Index·All Microsoft models
Microsoft
MicrosoftEfficientLIVE INDEX

Phi 4 Mini Instruct

text->text

Compact Phi-4 — 4B parameters tuned for cheap reasoning on commodity hardware.

Phi 4 Mini Instruct is a efficient AI model from Microsoft. It costs $0.080 per million input tokens and $0.350 per million output tokens (blended $0.161/M), with a 131K-token context window.

Profile inherited from upstream Phi-4 Mini ↗ — this is a hosted variant of the same open-weights model.

Released Feb 2025Modalities textOfficial model page ↗API docs ↗Pricing source ↗Compare with another model →Estimate monthly cost →
INPUT
$0.080/M
per million input tokens
OUTPUT
$0.350/M
per million output tokens
BLENDED 70/30
$0.161/M
default reference rate · how it's calculated →
CONTEXT
131K
131,072 tokens
What it's good at
  • Edge / consumer-GPU friendly
  • Open weights (MIT)
  • Strong math for 4B
Typical use cases
  • On-device reasoning
  • Cheap STEM tutoring
Benchmarks
vs. best public score
Scores inherited from Phi-4 Mini — this is a hosted variant of the same open-weights model, so the underlying benchmark scores are identical.
MMLU67%
Multitask academic knowledge across 57 subjects.
GPQA Diamond36%
Graduate-level science questions, "Google-proof".
MATH65%
High-school competition math problems.
HumanEval65%
Python function synthesis from docstrings.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources — treat as directional rather than authoritative.
More from Microsoft
See all 8 →
Phi-4
Frontier SLM · 16K ctx
in $0.070/Mout $0.280/M
Phi-4 Mini
Tiny · 128K ctx
in $0.040/Mout $0.160/M
Phi-3.5 Mini
Compact · 128K ctx
in $0.130/Mout $0.520/M
Phi-3.5 MoE
MoE · 128K ctx
in $0.160/Mout $0.640/M
Phi-3 Medium
Balanced · 128K ctx
in $0.170/Mout $0.680/M
Phi-3 Mini
Tiny · 128K ctx
in $0.130/Mout $0.520/M
Frequently asked questions

How much does Phi 4 Mini Instruct cost?

Phi 4 Mini Instruct costs $0.080 per million input tokens and $0.350 per million output tokens, for a blended reference rate of $0.161 per million tokens.

What is Phi 4 Mini Instruct's context window?

Phi 4 Mini Instruct supports up to 131K tokens of context (131,072 tokens).

What is Phi 4 Mini Instruct best for?

Phi 4 Mini Instruct is well suited to Edge / consumer-GPU friendly, Open weights (MIT) and Strong math for 4B.

Who makes Phi 4 Mini Instruct?

Phi 4 Mini Instruct is developed and served by Microsoft. It was released in Feb 2025.