Skip to content
Local AI Wizard
Your machine
Theme

Phi-4 family: which size to run

2 sizes, 3.8B–14B parameters · Microsoft

Checked

Sizes

Phi-4 sizes
ModelArchitecture
Phi-4 miniMicrosoft · MIT3.8BDense131,072No LiveBench result for this model in release 2026-06-251152.52.2
Phi-4Microsoft · MIT · Dec 12, 202414BDense16,384No LiveBench result for this model in release 2026-06-251913The repository labels its file Q4_K (llama.cpp's alias for Q4_K_M); the site's file ladder matches only the Q4_K_M label and its UD- variant8.2

Each size is its own model, with its own scores and memory needs. Q4 estimates are planning figures, not measurements; file sizes are published. The Mac or PC column is for a 32 GB or larger machine (8 GB reserve); machines under 32 GB reserve 25% of their memory (2 to 8 GB), so they need less.

The two cards describe models with different designs: the mini card says it “employed new architecture for efficiency, larger vocabulary for multilingual support”. Compare them as two models, not as one model at two sizes.

How to read this table

Q4 means compressed to about 4 bits per weight. GGUF (llama.cpp, Ollama, LM Studio) and MLX (Apple silicon) are file formats; the sizes shown are the published 4-bit files. A dash means no value is published or collected for that size; the reason is in the cell's accessible text.

Which size fits my machine

The largest size whose Q4 planning estimate fits what each machine offers. A planning estimate, not a measurement and not a speed.

Largest Phi-4 size per machine
MachineLargest Phi-4 size with enough memory
Mac mini M2 · 8 GB unified8Phi-4 mini5
Mac mini M4 · 16 GB unified16Phi-415
Mac mini M4 · 24 GB unified24Phi-417
Mac mini M4 · 32 GB unified32Phi-419
Mac mini M5 Pro · 64 GB unified56Phi-419
Mac Studio M5 Max · 128 GB unified104Phi-419
Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unified392Phi-419
RTX 4060 · 8 GB VRAM8Phi-4 mini5
RTX 3060 · 12 GB VRAM12Phi-4 mini5
RTX 5080 · 16 GB VRAM16Phi-413
RTX 4090 · 24 GB VRAM24Phi-413
RTX 5090 · 32 GB VRAM32Phi-413
DGX H100 · 8 × 80 GB640Phi-427

A dash means no size of this family fits that machine. The requirement is the size's Q4 estimate with that machine's own reserve: a graphics card reserves 2 GB, as in the sizes table; a Mac or PC reserves 25% of its memory (2 to 8 GB), so below 32 GB its figure is lower than the sizes table's. A server profile reserves 2 GB per graphics card, so its figure is higher.

Smallest machine for each size

The profile with the least memory that still has enough for the size's Q4 estimate, from the hardware catalog.

Smallest machine per Phi-4 size
SizeSmallest Mac or PC-memory profileSmallest graphics-card profile
Phi-4 miniMac mini 2020 · M1 · 8 GBRTX 3050 · 6 GB
Phi-4Mac mini 2020 · M1 · 16 GBRTX 4060 Ti · 16 GB

A dash means no profile of that kind in the catalog has enough memory for the size.

Memory by file precision

The same planning formula at three precisions (0.625, 1.125 and 2 bytes per parameter, plus working room and the system reserve). Estimates, not measurements.

Phi-4 memory estimate by precision
Size
Phi-4 mini1151481812
Phi-4191327214236

Family tree

Family, sizes and, for each size, the two 4-bit files the tables use (GGUF Q4_K_M and MLX 4-bit) when one was collected. Other precisions are not drawn here.

Phi-4 family: sizes and published filesSwipe sideways to see the files, or read the list below.
Phi-4 family: sizes and published filesFamilySizesFilesPhi-4Microsoftmini 3.8BGGUF Q4_K_M2.5 GBMLX 4-bit2.2 GB14BMLX 4-bit8.2 GB

The same tree as a list:

  • Phi-4 · Microsoft
    • mini 3.8B
      • GGUF Q4_K_M · 2.5 GB · unsloth/Phi-4-mini-instruct-GGUF, read Oct 23, 2026
      • MLX 4-bit · 2.2 GB · mlx-community/Phi-4-mini-instruct-4bit, read Oct 23, 2026
    • 14B
      • MLX 4-bit · 8.2 GB · mlx-community/phi-4-4bit, read Oct 23, 2026

Learn more

Sources

Where the family and its sizes are named, and where each number comes from.