What GPU do I need to run microsoft/Phi-4-multimodal-instruct?

5.6B parameters, published in BF16. View on Hugging Face

5.6B
Parameters
BF16
Native precision
Phi4MMForCausalLM
Architecture
automatic-speech-recognition
Pipeline

Phi-4-multimodal-instruct is published by microsoft on Hugging Face, with 258,928 downloads and 1,613 likes to date. It's a Phi4MMForCausalLM model built for automatic-speech-recognition, published natively in BF16.

VRAM required & cheapest live GPU fit

Required VRAM = weight size at each precision, plus a fixed overhead for activations and allocator fragmentation. Speech models don't build the same growing KV-cache a text model does. Memory scales primarily with input audio length. Full formula and assumptions: methodology.

PrecisionWeight sizeRequired VRAMCheapest live fitGPUs neededEst. $/hr (full fit)
BF1610.4 GB12.5 GBRTX A40001$0.167/hr
FP8 (quantized)5.2 GB6.2 GBRTX 5060 Ti1$0.180/hr
INT4 (quantized)2.6 GB3.1 GBRTX 30701$0.088/hr

A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.

INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.

Cheapest way to run Phi-4-multimodal-instruct at its published (BF16) precision: 1× RTX A4000, at $0.167/hr per GPU ($0.167/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.

Phi-4-multimodal-instruct: common questions

Does Phi-4-multimodal-instruct fit on a 16 GB GPU?

Yes. At BF16 it needs 12.5 GB of VRAM, so a 16 GB card holds it with 3.5 GB to spare. A 12 GB card is not enough for it at BF16.

What is the least VRAM Phi-4-multimodal-instruct can run in?

3.1 GB, at INT4 (quantized), which fits a 6 GB card, against 12.5 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.

Does quantizing Phi-4-multimodal-instruct lower the GPU bill?

Yes. At BF16 the cheapest live fit is one RTX A4000 at $0.167/hr. At INT4 (quantized) it drops to one RTX 3070 at $0.088/hr, provided a quantized checkpoint exists for it.

Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.

More Phi-4 models

All 7 Phi-4 models: VRAM and GPU requirements

Alternatives at this size

Other models for automatic-speech-recognition within about a third of Phi-4-multimodal-instruct's 5.6B parameters, from other model lines.

More on Phi-4-multimodal-instruct

Related reading: RTX A4000 pricing and specs, Whisper variants compared, and What AI inference is.

Your own GPU cloud accounts

One console for the GPU clouds you already rent from. Your keys stay on your machine.

$20 a month, flat

Connect your accounts

One console for the GPU clouds you already use.

Connect your accounts for $20 a month, flat. Your provider bills the GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.