PH

phi4 mm

PhiGGUF by Swicked86Updated 2026-03-20
Download on Hugging Face
Parameters3.84B
Context128K
Layers32
Architecturephi3
Smallest quant2.5 GB
Quick answer

phi4 mm needs about 4.1 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 3 quants, from 2.5 GB (Q4_K_M) to 7.7 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Phi-4 Multimodal Instruct – 3.84B parameter model with text, vision, and audio support

Phi-4 Multimodal Instruct is a 3.84B parameter multimodal foundation model. It processes text, images, and audio, generating text output. The model supports a 131072 token context length and is released under the MIT license.

Good at

  • Multimodal input handling
  • Large context length
  • Open source licensing

Limits

  • High VRAM requirement for full model
  • No further quantization beyond listed files
  • Requires mmproj file for vision

Best for

  • Multimodal chat applications
  • Vision‑based question answering
  • Audio transcription and translation

Who it’s for. Developers building multimodal AI assistants or research projects that need text, image, and audio processing in a single model. The model is suitable for those who can allocate sufficient GPU memory and are comfortable with open‑source licensing.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M2.5 GB 4.1 GB
Fits Get file →
Q8_04.1 GB 5.7 GB
Fits Get file →
F167.7 GB 9.3 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (2.49 GB file, 4.1 GB memory at 8 k context). For 12 GB VRAM, Q8_0 (4.08 GB file, 5.7 GB memory). For 16 GB VRAM or more, F16 (7.68 GB file, 9.3 GB memory). Estimates are approximate.

What GPU do you need for phi4 mm?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.7 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.7 GB in use).
  • 12 GB VRAM: the best quant that fits is F16 (7.7 GB file, about 9.3 GB in use).

Which phi4 mm quant should you choose?

Q4_K_M is the usual starting point for phi4 mm: it balances file size and quality well. Here it is a 2.5 GB file.

The Q8_0 file is 4.1 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturephi3
Parameters3.84B
Layers32
Trained context131,072 tokens
KV cache (fp16)128 KB per token

Source

Quantized bySwicked86
FamilyPhi
Licensemit
Downloads (HF)656
Last updated2026-03-20

Frequently asked questions about phi4 mm

How much VRAM does phi4 mm need?

About 4.1 GB at Q4_K_M with an 8K context window. That is the 2.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run phi4 mm on an 8 GB GPU?

Yes, with Q8_0: the 4.1 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of phi4 mm GGUF?

It depends on the quant: Q4_K_M is 2.5 GB, Q8_0 is 4.1 GB, F16 is 7.7 GB.

What context length does phi4 mm support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download phi4 mm GGUF?

The files are on Hugging Face in the Swicked86/phi4-mm-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does phi4 mm use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Q4_K_M on 8 GB VRAM?

Yes, Q4_K_M requires about 4.1 GB of VRAM at 8 k context, which fits within 8 GB.

Does this model support audio input?

Yes, it accepts audio inputs in English, Chinese, German, French, Italian, Japanese, Spanish, and Portuguese.

What license does the model use?

The model is released under the MIT license, shared by Microsoft and the community.

Other GGUFs of Phi-4-multimodal-instruct

Similar models

Browse all →

More Phi models

All Phi →

Other small models

All small models →