InterconnectsInterconnects Artifacts Hub
withVAILVAIL
All models

Phi-4-multimodal-instruct

byMicrosoftMicrosoft· 24 Feb 2025
General purposeMultimodal

An update to Phi, supporting text, images and audio as input.

Specs
Params5.6B
LicenseMIT
Tags
models/llmsmodels/llms/multi-modalmultilingual
Similarity · VAIL
VAIL
VAIL Fingerprint
0478:0592:082a:0b7e:1292:1bca:2545:62b0

Explore other models with behavioral similarity to Phi-4-multimodal-instruct.

Resources
Hugging Face — microsoft/Phi-4-multimodal-instructPaper / tech reportAppeared in Artifacts Log #8
How these metrics work
Capability · Artificial Analysis
AA Index
4.2
Adoption · Hugging Face
RAM @ 30d
Relative Adoption Metric: 0.36×. Below benchmark. Measured at 30 days.
Hugging Face Downloads
407.2K
last 30d
8.4M
all time
HF Likes
1.6K

Relative Adoption Metric contextualizes downloads against the model's size bucket.

Related Models

More from Microsoft

Phi-4-reasoning-vision-15B
MicrosoftMicrosoft04 Mar 2026
Phi-Ground-4B-7C
MicrosoftMicrosoft15 Aug 2025
Phi-4-mini-flash-reasoning
MicrosoftMicrosoft19 Jun 2025
Phi-4-reasoning-plus
MicrosoftMicrosoft30 Apr 2025

Similar Models

Gemma 3 27B IT
GoogleGoogle12 Mar 2025
multimodal27B params
gemma-4-E2B-it
GoogleGoogle02 Apr 2026
multimodal5B params
Step3-VL-10B
StepFunStepFun13 Jan 2026
multimodal10B params
Mistral Small 3.1 24B Instruct
MistralMistral11 Mar 2025
multimodal24B params