Outmake Models
Models

Qwen3 30B A3B 2507

Alibaba Qwen · July 2025

Get Hugging Face

A 31B, 3.4B active chat model from Alibaba Qwen.

Downloads
660K
Likes
374
Parameters
31B, 3.4B active
Context
256K
Licence
apache-2.0

Runs well on M4 · 16 GB

The UD-IQ1_M size from unsloth is the one to get: a 9.7 GB download that uses 10 GB of memory, and writes about 9.9 words a second.

Qwen3 30B A3B 2507 runs well from a Mac with an M1 and 16 GB of memory. 56 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
UD-TQ1_08.1 GB8.7 GBestimatedOpen
UD-IQ1_S9.0 GB9.6 GBestimatedOpen
UD-IQ1_M9.7 GB10 GBestimatedOpen
UD-IQ2_XXS10 GB11 GBestimatedOpen
UD-IQ2_M11 GB11 GBestimatedOpen
Q2_K11 GB12 GBestimatedOpen
Q2_K_L11 GB12 GBestimatedOpen
UD-Q2_K_XL12 GB12 GBestimatedOpen
UD-IQ3_XXS13 GB13 GBestimatedOpen
Q3_K_S13 GB14 GBestimatedOpen
UD-Q3_K_XL14 GB14 GBestimatedOpen
Q3_K_M15 GB15 GBestimatedOpen
IQ4_XS16 GB17 GBestimatedOpen
IQ4_NL17 GB18 GBestimatedOpen
Q4_017 GB18 GBestimatedOpen
Q4_K_S17 GB18 GBestimatedOpen
UD-Q4_K_XL18 GB18 GBestimatedOpen
Q4_K_M19 GB19 GBestimatedOpen
Q4_119 GB20 GBestimatedOpen
Q5_K_S21 GB22 GBestimatedOpen
Q5_K_M22 GB22 GBestimatedOpen
UD-Q5_K_XL22 GB22 GBestimatedOpen
Q6_K25 GB26 GBestimatedOpen
UD-Q6_K_XL26 GB27 GBestimatedOpen
Q8_032 GB33 GBestimatedOpen
UD-Q8_K_XL36 GB37 GBestimatedOpen
BF1661 GB62 GBestimatedOpen

Run It Anywhere

Qwen3 30B A3B 2507 is an open model, so any app that runs GGUF files can run it. These take the UD-IQ1_M size from unsloth straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF:UD-IQ1_M
  • llama.cppMac, Windows, Linuxllama-server -hf unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF:UD-IQ1_M
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF:UD-IQ1_M
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 11 GB
8 GBNone worth runningneeds 11 GB
10 GBNone worth runningneeds 11 GB
11 GBNone worth runningneeds 11 GB
12 GBUD-IQ2_M11 GB11 GB
16 GBQ3_K_M15 GB15 GB
20 GBQ4_K_M19 GB19 GB
24 GBUD-Q5_K_XL22 GB22 GB
32 GBUD-Q6_K_XL26 GB27 GB
48 GBUD-Q8_K_XL36 GB37 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 GBUD-IQ1_MRuns well10 GB15
M6 · 24 GBQ3_K_MRuns well16 GB13
M6 · 32 GBQ4_1Runs well21 GB14
M5 · 16 GBUD-IQ1_MRuns well10 GB14
M5 · 24 GBQ3_K_MRuns well16 GB12
M5 · 32 GBQ4_1Runs well21 GB13
M5 Pro · 24 GBQ3_K_MRuns well16 GB24
M5 Pro · 48 GBUD-Q5_K_XLRuns great23 GB21
M5 Pro · 64 GBUD-Q6_K_XLRuns great30 GB19
M5 Max · 36 GBIQ4_XSRuns great17 GB33
M5 Max · 48 GBUD-Q5_K_XLRuns great23 GB38
M5 Max · 64 GBUD-Q6_K_XLRuns great30 GB35
M5 Max · 128 GBQ8_0Runs great47 GB38
M5 Ultra · 96 GBQ8_0Runs great40 GB66
M5 Ultra · 256 GBQ8_0Runs great47 GB66
M5 Ultra · 512 GBQ8_0Runs great47 GB66
M4 · 16 GBUD-IQ1_MRuns well10 GB9.9
M4 · 24 GBQ3_K_MRuns well16 GB9
M4 · 32 GBQ4_1Runs well21 GB9.3
M4 Pro · 24 GBQ3_K_MRuns well16 GB17
M4 Pro · 48 GBUD-Q5_K_XLRuns great23 GB15
M4 Pro · 64 GBUD-Q6_K_XLRuns great30 GB14
M4 Max · 36 GBIQ4_XSRuns great17 GB22
M4 Max · 48 GBUD-Q5_K_XLRuns great23 GB25
M4 Max · 64 GBUD-Q6_K_XLRuns great30 GB23
M4 Max · 96 GBQ8_0Runs great40 GB20
M4 Max · 128 GBQ8_0Runs great47 GB25
M3 · 8 GBIQ2_XXSToo big23 GB
M3 · 16 GBUD-IQ1_MRuns well10 GB8.7
M3 · 24 GBQ3_K_MRuns well16 GB7.8
M3 Pro · 18 GBQ2_K_LRuns well12 GB12
M3 Pro · 36 GBIQ4_XSRuns great17 GB11
M3 Max · 36 GBIQ4_XSRuns great17 GB19
M3 Max · 48 GBUD-Q5_K_XLRuns great23 GB21
M3 Max · 64 GBUD-Q6_K_XLRuns great30 GB20
M3 Max · 96 GBQ8_0Runs great40 GB17
M3 Max · 128 GBQ8_0Runs great47 GB21
M3 Ultra · 96 GBQ8_0Runs great40 GB34
M3 Ultra · 256 GBQ8_0Runs great47 GB34
M3 Ultra · 512 GBQ8_0Runs great47 GB34
M2 · 8 GBIQ2_XXSToo big23 GB
M2 · 16 GBUD-IQ1_MRuns well10 GB8.7
M2 · 24 GBQ3_K_MRuns well16 GB7.8
M2 Pro · 16 GBUD-IQ1_MRuns well10 GB15
M2 Pro · 32 GBQ4_1Runs well21 GB15
M2 Max · 32 GBQ4_1Runs well21 GB25
M2 Max · 64 GBUD-Q6_K_XLRuns great30 GB20
M2 Max · 96 GBQ8_0Runs great40 GB21
M2 Ultra · 64 GBUD-Q6_K_XLRuns great30 GB31
M2 Ultra · 128 GBQ8_0Runs great47 GB33
M2 Ultra · 192 GBQ8_0Runs great47 GB33
M1 · 8 GBIQ2_XXSToo big23 GB
M1 · 16 GBUD-IQ1_MRuns well10 GB5.8
M1 Pro · 16 GBUD-IQ1_MRuns well10 GB14
M1 Pro · 32 GBQ4_1Runs well21 GB13
M1 Max · 32 GBQ4_1Runs well21 GB21
M1 Max · 64 GBUD-Q6_K_XLRuns great30 GB17
M1 Ultra · 64 GBUD-Q6_K_XLRuns great30 GB24
M1 Ultra · 128 GBQ8_0Runs great47 GB26

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

Packaged by

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

Information

Maker
Alibaba Qwen
Released
July 2025
Parameters
31B, 3.4B active
Longest chat
256K tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
659,882
Likes
374
Sizes
27, 8.1 to 61 GB
Kind
Maker's release

Badge for the model card

Shows on Qwen3 30B A3B 2507'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