Phi-4-Reasoning-Vision-15B

Phi-4-Reasoning-Vision-15B.

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Microsoft Mar 2026

Open weights Small

No Epoch Capabilities Index score yet

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Benchmark	Score		Rank	Source
HallusionBench	64.4%		#14 of 31	Lab model card

Benchmark	Score		Rank	Source
PHYSICS-Verified	14.6%		#17 of 19	Official leaderboard

Benchmark	Score		Rank	Source
ScreenSpot-v2	88.2%		#12 of 14	Lab model card

Benchmark	Score	Rank	Source
MathVista	75.2%	#35 of 70	Lab model card
ChartQA	83.3%	#23 of 37	Lab model card
AI2D	84.8%	#39 of 52	Lab model card
MMStar	64.5%	#31 of 36	Lab model card
MMMU	54.3%	#93 of 103	Lab model card
OCRBench	76%	#32 of 34	Lab model card
MATH-Vision (MathVision)	36.2%	#50 of 51	Lab model card
MathVerse	44.9%	#12 of 12	Lab model card

Benchmark	Score		Rank	Source
Harmful content defect rate (Azure AI Evaluation SDK, Microsoft)	1.40%		#1 of 3	Lab model card

The benchmarks the lab chose to headline, with the scores it gave.

Every result in the longer documents the lab published with it. 1 of them isn't in the launch post.