ModelsCompareBest forBenchmarksStatusPricingAPI

Phi-4 Reasoning

Microsoft·Pending·Released 2025-04-30text
Context32.768k
Max output32.768k
Input / 1M-
Output / 1M-
Cached in-
Benchmarks4

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

AIME 2025#13/1562.9% ↗
GPQA Diamond#58/7065.8% ↗
HumanEval+#2/492.9% ↗
MMLU-Pro#22/2974.3% ↗

Compare

vs GPT-6 Astravs GPT-5.6 Solvs GPT-5.6 Terravs GPT-5.6 Lunavs Claude Fable 5.1vs Claude Opus 5vs Claude Sonnet 5vs Claude Haiku 4.5

More from Microsoft

MAI-Thinking-1Phi-4Phi-4 Mini ReasoningPhi-4 Multimodal Instruct

Access Phi-4 Reasoning and every other model through one endpoint with automatic failover: Respan gateway.