ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

Phi-4vsPhi-4 Mini Reasoning

Phi-4 leads 1 of 1 shared benchmark. Phi-4 Mini Reasoning has the larger context window (128K tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
1 – 0
Phi-4 leads
Cheaper per token
list price, input + output
Larger context
Phi-4 Mini Reasoning
128K tokens
Provider uptime (30d)
Microsoft
Phi-4
Microsoft · released 2024-12-12
text
Context
16K
Max out
16K
Input /1M
Output /1M
Cached /1M
Scores
11 · 7 core
Phi-4 Mini Reasoning
Microsoft · released 2025-04-30
text
Context
128K
Max out
128K
Input /1M
Output /1M
Cached /1M
Scores
3 · 2 core
Quality

Benchmark matrix

BenchmarkPhi-4Phi-4 Mini ReasoningΔEdge
Reported by both · 1
GPQA Diamond56.1% 52% +4.1 ptPhi-4
Only Phi-4 reports · 9
MMLU84.8% not reported
MMLU-Pro70.4% not reported
HumanEval82.6% not reported
LMArena Elo1216.6 not reported
ArenaHard75.4% not reported
HumanEval+82.8% not reported
IFEval63% not reported
MATH80.4% not reported
OTIS Mock AIME 2024-2025 (Epoch AI run)13.8% epochnot reported
Only Phi-4 Mini Reasoning reports · 2
AIME 2024not reported57.5%
MATH-500not reported94.6%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Phi-4 minus Phi-4 Mini Reasoning in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecPhi-4Phi-4 Mini ReasoningEdge
Context window16K tokens128K tokensPhi-4 Mini Reasoning
Max output16K tokens128K tokensPhi-4 Mini Reasoning
Input price / 1M
Output price / 1M
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
ModalitiestexttextTie
Released2024-12-122025-04-30
Cited benchmark scores113
Reliability

Microsoft status

All providers
More matchups

Phi-4 Mini Reasoning vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run Phi-4 and Phi-4 Mini Reasoning on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.