Outmake Models
Models

Qwen3 14B

Alibaba Qwen · April 2025

Get Hugging Face

A 15B chat model from Alibaba Qwen.

Downloads
619K
Likes
405
Parameters
15B
Context
40K
Licence
apache-2.0

Runs well on M4 · 16 GB

The Q4_K_M size from Qwen is the one to get: a 9.0 GB download that uses 9.9 GB of memory, and writes about 4.8 words a second.

Qwen3 14B 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 Qwen/Qwen3-14B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_K_M9.0 GB9.9 GBestimatedOpen
Q5_010 GB11 GBestimatedOpen
Q5_K_M11 GB11 GBestimatedOpen
Q6_K12 GB13 GBestimatedOpen
Q8_016 GB17 GBestimatedOpen

Run It Anywhere

Qwen3 14B is an open model, so any app that runs GGUF files can run it. These take the Q4_K_M size from Qwen straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/Qwen/Qwen3-14B-GGUF:Q4_K_M
  • llama.cppMac, Windows, Linuxllama-server -hf Qwen/Qwen3-14B-GGUF:Q4_K_M
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/Qwen/Qwen3-14B-GGUF:Q4_K_M
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 9.9 GB
8 GBNone worth runningneeds 9.9 GB
10 GBNone worth runningneeds 9.9 GB
11 GBQ4_K_M9.0 GB9.9 GB
12 GBQ5_K_M11 GB11 GB
16 GBQ6_K12 GB13 GB
20 GBQ8_016 GB17 GB
24 GBQ8_016 GB17 GB
32 GBQ8_016 GB17 GB
48 GBQ8_016 GB17 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 GBQ4_K_MRuns well9.9 GB6.7
M6 · 24 GBQ5_K_MRuns great11 GB6.1
M6 · 32 GBUD-Q6_K_XLRuns great15 GB5.2
M5 · 16 GBQ4_K_MRuns well9.9 GB6.1
M5 · 24 GBQ5_K_MRuns great11 GB5.5
M5 · 32 GBUD-Q6_K_XLRuns great15 GB4.7
M5 Pro · 24 GBQ5_K_MRuns great11 GB11
M5 Pro · 48 GBQ8_0Runs great19 GB9.8
M5 Pro · 64 GBQ8_0Runs great19 GB9.8
M5 Max · 36 GBQ8_0Runs great17 GB14
M5 Max · 48 GBQ8_0Runs great19 GB19
M5 Max · 64 GBQ8_0Runs great19 GB19
M5 Max · 128 GBQ8_0Runs great19 GB19
M5 Ultra · 96 GBQ8_0Runs great19 GB37
M5 Ultra · 256 GBQ8_0Runs great19 GB37
M5 Ultra · 512 GBQ8_0Runs great19 GB37
M4 · 16 GBQ4_K_MRuns well9.9 GB4.8
M4 · 24 GBQ5_K_MRuns great11 GB4.3
M4 · 32 GBUD-Q6_K_XLRuns great15 GB3.7
M4 Pro · 24 GBQ5_K_MRuns great11 GB9.4
M4 Pro · 48 GBQ8_0Runs great19 GB8.2
M4 Pro · 64 GBQ8_0Runs great19 GB8.2
M4 Max · 36 GBQ8_0Runs great17 GB12
M4 Max · 48 GBQ8_0Runs great19 GB15
M4 Max · 64 GBQ8_0Runs great19 GB15
M4 Max · 96 GBQ8_0Runs great19 GB12
M4 Max · 128 GBQ8_0Runs great19 GB15
M3 · 8 GBUD-IQ1_MRuns well4.9 GB6.1
M3 · 16 GBQ4_K_MRuns well9.9 GB4
M3 · 24 GBQ5_K_MRuns great11 GB3.7
M3 Pro · 18 GBUD-Q3_K_XLRuns great8.5 GB6.6
M3 Pro · 36 GBQ8_0Runs great17 GB4.7
M3 Max · 36 GBQ8_0Runs great17 GB9.1
M3 Max · 48 GBQ8_0Runs great19 GB12
M3 Max · 64 GBQ8_0Runs great19 GB12
M3 Max · 96 GBQ8_0Runs great19 GB9.1
M3 Max · 128 GBQ8_0Runs great19 GB12
M3 Ultra · 96 GBQ8_0Runs great19 GB22
M3 Ultra · 256 GBQ8_0Runs great19 GB22
M3 Ultra · 512 GBQ8_0Runs great19 GB22
M2 · 8 GBUD-IQ1_MRuns well4.9 GB6.1
M2 · 16 GBQ4_K_MRuns well9.9 GB4
M2 · 24 GBQ5_K_MRuns great11 GB3.7
M2 Pro · 16 GBQ4_K_MRuns well9.9 GB7.8
M2 Pro · 32 GBUD-Q6_K_XLRuns great15 GB6.1
M2 Max · 32 GBUD-Q6_K_XLRuns great15 GB12
M2 Max · 64 GBQ8_0Runs great19 GB12
M2 Max · 96 GBQ8_0Runs great19 GB12
M2 Ultra · 64 GBQ8_0Runs great19 GB22
M2 Ultra · 128 GBQ8_0Runs great19 GB22
M2 Ultra · 192 GBQ8_0Runs great19 GB22
M1 · 8 GBUD-IQ1_MRuns well4.9 GB4.1
M1 · 16 GBQ4_K_MRuns well9.9 GB2.7
M1 Pro · 16 GBQ4_K_MRuns well9.9 GB7.5
M1 Pro · 32 GBUD-Q6_K_XLRuns great15 GB5.9
M1 Max · 32 GBUD-Q6_K_XLRuns great15 GB11
M1 Max · 64 GBQ8_0Runs great19 GB11
M1 Ultra · 64 GBQ8_0Runs great19 GB20
M1 Ultra · 128 GBQ8_0Runs great19 GB20

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

Packaged by

10 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.

Information

Maker
Alibaba Qwen
Released
April 2025
Parameters
15B
Longest chat
40K tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
619,349
Likes
405
Sizes
5, 9.0 to 16 GB
Kind
Maker's release

Badge for the model card

Shows on Qwen3 14B'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