Outmake Models
Models

CreativeWriter Llama3.2 3B

theprint · November 2025

Get Hugging Face

A 3.2B chat model from theprint.

Downloads
47K
Parameters
3.2B
Context
128K
Licence
apache-2.0

Runs great on M4 · 16 GB

The Q8_0 size from theprint is the one to get: a 3.4 GB download that uses 5.7 GB of memory, and writes about 11 words a second.

CreativeWriter Llama3.2 3B runs well from a Mac with an M1 and 8 GB of memory. 59 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From theprint/CreativeWriter-Llama3.2-3B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q2_K1.4 GB2.0 GBestimatedOpen
Q3_K_S1.5 GB2.2 GBestimatedOpen
Q3_K_M1.7 GB2.3 GBestimatedOpen
Q3_K_L1.8 GB2.5 GBestimatedOpen
IQ4_XS1.8 GB2.5 GBestimatedOpen
Q4_01.9 GB2.6 GBestimatedOpen
Q4_K_M2.0 GB2.7 GBestimatedOpen
Q4_12.1 GB2.7 GBestimatedOpen
Q5_K_M2.3 GB3.0 GBestimatedOpen
Q6_K2.6 GB3.3 GBestimatedOpen
Q8_03.4 GB4.1 GBestimatedOpen
F166.4 GB7.1 GBestimatedOpen

Run It Anywhere

CreativeWriter Llama3.2 3B is an open model, so any app that runs GGUF files can run it. These take the Q8_0 size from theprint straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/theprint/CreativeWriter-Llama3.2-3B-GGUF:Q8_0
  • llama.cppMac, Windows, Linuxllama-server -hf theprint/CreativeWriter-Llama3.2-3B-GGUF:Q8_0
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/theprint/CreativeWriter-Llama3.2-3B-GGUF:Q8_0
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBQ8_03.4 GB4.1 GB
8 GBF166.4 GB7.1 GB
10 GBF166.4 GB7.1 GB
11 GBF166.4 GB7.1 GB
12 GBF166.4 GB7.1 GB
16 GBF166.4 GB7.1 GB
20 GBF166.4 GB7.1 GB
24 GBF166.4 GB7.1 GB
32 GBF166.4 GB7.1 GB
48 GBF166.4 GB7.1 GB

The largest size, of those that keep three bits a weight or more, whose memory at an 8K chat is under 92% of the card's; the rest is the desktop's. Estimated from the files, the same way as for a Mac; speed on a PC is not estimated yet. A size that doesn't fit the card still runs, slower, with part of it in ordinary memory.

On every Mac
MacSizeFitMemoryWrites
M6 · 16 GBQ8_0Runs great5.7 GB16
M6 · 24 GBQ8_0Runs great7.9 GB16
M6 · 32 GBQ8_0Runs great12 GB16
M5 · 16 GBQ8_0Runs great5.7 GB14
M5 · 24 GBQ8_0Runs great7.9 GB14
M5 · 32 GBQ8_0Runs great12 GB14
M5 Pro · 24 GBQ8_0Runs great7.9 GB28
M5 Pro · 48 GBQ8_0Runs great12 GB28
M5 Pro · 64 GBQ8_0Runs great12 GB28
M5 Max · 36 GBQ8_0Runs great12 GB41
M5 Max · 48 GBQ8_0Runs great12 GB54
M5 Max · 64 GBQ8_0Runs great12 GB54
M5 Max · 128 GBQ8_0Runs great12 GB54
M5 Ultra · 96 GBQ8_0Runs great12 GB100
M5 Ultra · 256 GBQ8_0Runs great12 GB100
M5 Ultra · 512 GBQ8_0Runs great12 GB100
M4 · 16 GBQ8_0Runs great5.7 GB11
M4 · 24 GBQ8_0Runs great7.9 GB11
M4 · 32 GBQ8_0Runs great12 GB11
M4 Pro · 24 GBQ8_0Runs great7.9 GB23
M4 Pro · 48 GBQ8_0Runs great12 GB23
M4 Pro · 64 GBQ8_0Runs great12 GB23
M4 Max · 36 GBQ8_0Runs great12 GB32
M4 Max · 48 GBQ8_0Runs great12 GB40
M4 Max · 64 GBQ8_0Runs great12 GB40
M4 Max · 96 GBQ8_0Runs great12 GB32
M4 Max · 128 GBQ8_0Runs great12 GB40
M3 · 8 GBQ6_KRuns great3.8 GB8.9
M3 · 16 GBQ8_0Runs great5.7 GB9.1
M3 · 24 GBQ8_0Runs great7.9 GB9.1
M3 Pro · 18 GBQ8_0Runs great7.9 GB13
M3 Pro · 36 GBQ8_0Runs great12 GB13
M3 Max · 36 GBQ8_0Runs great12 GB25
M3 Max · 48 GBQ8_0Runs great12 GB32
M3 Max · 64 GBQ8_0Runs great12 GB32
M3 Max · 96 GBQ8_0Runs great12 GB25
M3 Max · 128 GBQ8_0Runs great12 GB32
M3 Ultra · 96 GBQ8_0Runs great12 GB55
M3 Ultra · 256 GBQ8_0Runs great12 GB55
M3 Ultra · 512 GBQ8_0Runs great12 GB55
M2 · 8 GBQ6_KRuns great3.8 GB8.9
M2 · 16 GBQ8_0Runs great5.7 GB9.1
M2 · 24 GBQ8_0Runs great7.9 GB9.1
M2 Pro · 16 GBQ8_0Runs great5.7 GB17
M2 Pro · 32 GBQ8_0Runs great12 GB17
M2 Max · 32 GBQ8_0Runs great12 GB32
M2 Max · 64 GBQ8_0Runs great12 GB32
M2 Max · 96 GBQ8_0Runs great12 GB32
M2 Ultra · 64 GBQ8_0Runs great12 GB55
M2 Ultra · 128 GBQ8_0Runs great12 GB55
M2 Ultra · 192 GBQ8_0Runs great12 GB55
M1 · 8 GBQ6_KRuns great3.8 GB6
M1 · 16 GBQ8_0Runs great5.7 GB6.1
M1 Pro · 16 GBQ8_0Runs great5.7 GB16
M1 Pro · 32 GBQ8_0Runs great12 GB16
M1 Max · 32 GBQ8_0Runs great12 GB29
M1 Max · 64 GBQ8_0Runs great12 GB29
M1 Ultra · 64 GBQ8_0Runs great12 GB47
M1 Ultra · 128 GBQ8_0Runs great12 GB47

Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.

Information

Maker
theprint
Released
November 2025
Parameters
3.2B
Longest chat
128K tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
47,370
Sizes
12, 1.4 to 6.4 GB
Kind
Maker's release

Badge for the model card

Shows on CreativeWriter Llama3.2 3B's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

Files, downloads and likes read from Hugging Face on November 5, 2026 and refreshed every morning. Memory and speed are estimates until someone running Outmake measures this model on that Mac.

Your Computer

Every model is judged for this Mac. Both are under About This Mac, in the Apple menu.

Every Mac · Every graphics card

Civitai

Open in OutmakeHugging Face

New to Outmake? Download the free beta