Skip to content
Local AI Wizard
Your machine
Theme

DeepSeek V4 family: which size to run

2 sizes, 284B–1.6T parameters · DeepSeek

Checked

Sizes

DeepSeek V4 sizes
ModelArchitecture
DeepSeek V4 FlashDeepSeek · MIT · Apr 24, 2026284BMoE · 13B active1,048,57665.5221215No GGUF Q4_K_M file (plain or UD-) in the repository collected160.1
DeepSeek V4 ProDeepSeek · MIT · Apr 24, 20261.6TMoE · 49B active1,048,57671.61,2081,202No file facts collected for this modelNo file facts collected for this model

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 publisher lists V4-Pro 1.6T (49B active), V4-Flash 284B (13B active).

The two sizes listed are the preview checkpoints. DeepSeek later published official releases of both (below), each with its own scores. All are very large mixture-of-experts models.

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. Smaller machines are left out; the largest one that fits no size is kept to show the limit.

Largest DeepSeek V4 size per machine
MachineLargest DeepSeek V4 size with enough memory
Mac Studio M5 Max · 128 GB unified104No size of this family has a Q4 estimate within what this machine offersNo size fits, so there is no estimate to compare
Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unified392DeepSeek V4 Flash221
RTX 5090 · 32 GB VRAM32No size of this family has a Q4 estimate within what this machine offersNo size fits, so there is no estimate to compare
DGX H100 · 8 × 80 GB640DeepSeek V4 Flash229

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 DeepSeek V4 size
SizeSmallest Mac or PC-memory profileSmallest graphics-card profileServer-class profile
DeepSeek V4 FlashMac Studio 2026 · M5 Ultra · 512 GBNo graphics-card profile in the catalog has enough memory for this sizeDGX H100 · 8 × 80 GB · 640 GB
DeepSeek V4 ProNo Mac, laptop or PC profile in the catalog has enough memory for this sizeNo graphics-card profile in the catalog has enough memory for this sizeEven the largest profile in the catalog (DGX H100 · 8 × 80 GB, 640 GB aggregated) is below this planning estimate

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.

DeepSeek V4 memory estimate by precision
Size
DeepSeek V4 Flash221215392386690684
DeepSeek V4 Pro1,2081,2022,1682,1623,8483,842

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.

DeepSeek V4 family: sizes and published filesSwipe sideways to see the files, or read the list below.
DeepSeek V4 family: sizes and published filesFamilySizesFilesDeepSeek V4DeepSeekFlash 284B13B activeMLX 4-bit160.1 GBPro 1.6T49B activeNo files collected

The same tree as a list:

  • DeepSeek V4 · DeepSeek
    • Flash 284B · 13B active
      • MLX 4-bit · 160.1 GB · mlx-community/deepseek-ai-DeepSeek-V4-Flash-4bit, read Oct 23, 2026
    • Pro 1.6T · 49B active
      • No files collected

Learn more

Sources

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