Phi 4 Multimodal Instruct vs Phi 4 R V 15B

Microsoft · released Feb 26, 2025

Wins 1 of 1 area

Images and charts

Microsoft · released Mar 4, 2026

Wins 0 of 1 area

—

Phi 4 Multimodal Instruct is the stronger all-rounder.

Scores updated Aug 23, 2026 · 24 tests both models report · How we compare

Where each one wins

Tests won in each of the one area where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.

Images and chartsUnderstanding pictures, charts and video30Phi 4 Multimodal Instruct3 of 4 tests · 1 tie

The biggest differences

The tests each model wins by the widest margin, up to three each. Scores are out of 100.

Where Phi 4 Multimodal Instruct pulls ahead

Reads text in imagesOCRBench+10.5points aheadPhi 4 Multimodal Instruct84.4Phi 4 R V 15B73.9
Visual perception tasks people solve at a glanceBLINK+3.5points aheadPhi 4 Multimodal Instruct61.3Phi 4 R V 15B57.8
Questions that truly need the image to answerMMStar+1.6points aheadPhi 4 Multimodal Instruct61.2Phi 4 R V 15B59.6

Where Phi 4 R V 15B pulls ahead

No clear win on a test scored out of 100.

Every test, side by side

All 24 tests both models report. The winning score is in its model's colour; marks a score checked independently.

Images and chartsPhi 4 Multimodal Instruct

OCRBenchPhi 4 Multimodal Instruct by 10.584.473.9+10.5
BLINKPhi 4 Multimodal Instruct by 3.561.357.8+3.5
MMStarPhi 4 Multimodal Instruct by 1.661.259.6+1.6
Video-MMEtie5555.3tie

Full images and charts ranking

Other results20 tests, not counted

Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.

TextVQA-valPhi 4 R V 15B by 36.239.976.1+36.2
MME-RealWorldPhi 4 R V 15B by 25.332.557.8+25.3
CV-BenchPhi 4 R V 15B by 24.257.181.3+24.2
Timelens-ActivityNetPhi 4 R V 15B by 21223+21
MMBench-EN-devPhi 4 R V 15B by 18.465.884.2+18.4
MME-PerceptionPhi 4 R V 15B by 180 rating points14101590+180 rating
DocVQA-valPhi 4 Multimodal Instruct by 16.692.876.2+16.6
InfoVQA (val)Phi 4 Multimodal Instruct by 16.471.855.4+16.4
NextQAPhi 4 R V 15B by 14.954.169+14.9
MultiDocVQA-valPhi 4 R V 15B by 11.646.858.4+11.6
SATPhi 4 R V 15B by 11.455.366.7+11.4
LongVideoBenchPhi 4 R V 15B by 10.141.151.2+10.1
LVBenchPhi 4 R V 15B by 9.125.334.4+9.1
MLVU-devPhi 4 R V 15B by 7.644.251.8+7.6
MMBench-CN-devPhi 4 R V 15B by 4.375.279.5+4.3
MV-BenchPhi 4 R V 15B by 4.344.949.2+4.3
ChartQAPhi 4 R V 15B by 281.483.4+2
VSI-BenchPhi 4 R V 15B by 1.424.125.5+1.4
AI2D w/ MaskPhi 4 R V 15B by 1.181.882.9+1.1
Ref-DAVIS17tie3.12.1tie

Questions people ask

Which is better, Phi 4 Multimodal Instruct or Phi 4 R V 15B?

Phi 4 Multimodal Instruct wins the one area where both have results: images and charts. Phi 4 R V 15B wins none.

How do you compare the two?

We use the 24 benchmark tests both models have published scores on. The verdict counts the 4 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.