Outmake Models
Models

Aya Expanse 32B

Cohere · October 2024

Get Hugging Face

A 32B chat model from Cohere.

Rating
2.7
Downloads
1.7K
Likes
24
Parameters
32B
Context
8K
Licence
cc-by-nc-4.0

Tight fit on M4 · 16 GB

The IQ2_S size from bartowski is the one to get: a 11 GB download that uses 12 GB of memory, and writes about 4.3 words a second.

Ratings

2.7 out of 5, across people's blind votes on its answers and the tests below.

Text
1267
Coding
1292

LM Arena ratings of October 28, 2026 (CC BY 4.0): people vote between two answers without knowing which model wrote which. The bar ends at 1488, the best any open model has. The Arena tests the full model; a small size on a Mac answers a little worse.

KnowledgeMMLU-Pro
45.4%

MMLU-Pro: graduate-level questions in 14 subjects. Scores are the ones filed on the model's Hugging Face page or the board's own table, read 2026-11-05; some come from the maker.

Aya Expanse 32B runs well from a Mac with an M3 Pro and 18 GB of memory. 48 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From bartowski/aya-expanse-32b-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
IQ2_XS10 GB11 GBestimatedOpen
IQ2_S11 GB12 GBestimatedOpen
IQ2_M12 GB12 GBestimatedOpen
Q2_K13 GB14 GBestimatedOpen
Q2_K_L13 GB14 GBestimatedOpen
IQ3_XS14 GB15 GBestimatedOpen
Q3_K_S15 GB16 GBestimatedOpen
IQ3_M15 GB16 GBestimatedOpen
Q3_K_M16 GB17 GBestimatedOpen
Q3_K_L18 GB18 GBestimatedOpen
IQ4_XS18 GB19 GBestimatedOpen
Q3_K_XL18 GB19 GBestimatedOpen
Q4_019 GB20 GBestimatedOpen
Q4_K_S19 GB20 GBestimatedOpen
Q4_K_M20 GB21 GBestimatedOpen
Q4_K_L20 GB21 GBestimatedOpen
Q5_K_S22 GB23 GBestimatedOpen
Q5_K_M23 GB24 GBestimatedOpen
Q5_K_L24 GB24 GBestimatedOpen
Q6_K27 GB27 GBestimatedOpen
Q6_K_L27 GB28 GBestimatedOpen
Q8_034 GB35 GBestimatedOpen
F1665 GB65 GBestimatedOpen

Run It Anywhere

Aya Expanse 32B is an open model, so any app that runs GGUF files can run it. These take the IQ2_S size from bartowski straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/bartowski/aya-expanse-32b-GGUF:IQ2_S
  • llama.cppMac, Windows, Linuxllama-server -hf bartowski/aya-expanse-32b-GGUF:IQ2_S
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/aya-expanse-32b-GGUF:IQ2_S
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 15 GB
8 GBNone worth runningneeds 15 GB
10 GBNone worth runningneeds 15 GB
11 GBNone worth runningneeds 15 GB
12 GBNone worth runningneeds 15 GB
16 GBQ3_K_S15 GB16 GB
20 GBQ4_K_S19 GB20 GB
24 GBQ5_K_S22 GB23 GB
32 GBQ6_K_L27 GB28 GB
48 GBQ8_034 GB35 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 GBIQ2_STight fit12 GB6
M6 · 24 GBQ3_K_SRuns well16 GB4.8
M6 · 32 GBQ4_K_LRuns well21 GB3.8
M5 · 16 GBIQ2_STight fit12 GB5.4
M5 · 24 GBQ3_K_SRuns well16 GB4.4
M5 · 32 GBQ4_K_LRuns well21 GB3.4
M5 Pro · 24 GBQ3_K_SRuns well16 GB8.6
M5 Pro · 48 GBQ4_K_LRuns great21 GB6.8
M5 Pro · 64 GBQ6_K_LRuns great28 GB5.6
M5 Max · 36 GBQ3_K_MRuns great17 GB12
M5 Max · 48 GBQ4_K_LRuns great21 GB13
M5 Max · 64 GBQ6_K_LRuns great28 GB11
M5 Max · 128 GBQ8_0Runs great35 GB10
M5 Ultra · 96 GBQ8_0Runs great35 GB20
M5 Ultra · 256 GBQ8_0Runs great35 GB20
M5 Ultra · 512 GBQ8_0Runs great35 GB20
M4 · 16 GBIQ2_STight fit12 GB4.3
M4 · 24 GBQ3_K_SRuns well16 GB3.5
M4 · 32 GBQ4_K_LRuns well21 GB2.7
M4 Pro · 24 GBQ3_K_SRuns well16 GB7.5
M4 Pro · 48 GBQ4_K_LRuns great21 GB6
M4 Pro · 64 GBQ6_K_LRuns great28 GB4.8
M4 Max · 36 GBQ3_K_MRuns great17 GB10
M4 Max · 48 GBQ4_K_LRuns great21 GB11
M4 Max · 64 GBQ6_K_LRuns great28 GB9.2
M4 Max · 96 GBQ8_0Runs great35 GB6.6
M4 Max · 128 GBQ8_0Runs great35 GB8.7
M3 · 8 GBIQ2_XSToo big11 GB
M3 · 16 GBIQ2_STight fit12 GB3.6
M3 · 24 GBQ3_K_SRuns well16 GB2.9
M3 Pro · 18 GBIQ2_SRuns well12 GB5.3
M3 Pro · 36 GBQ3_K_MRuns great17 GB4
M3 Max · 36 GBQ3_K_MRuns great17 GB7.8
M3 Max · 48 GBQ4_K_LRuns great21 GB8.6
M3 Max · 64 GBQ6_K_LRuns great28 GB7
M3 Max · 96 GBQ8_0Runs great35 GB4.9
M3 Max · 128 GBQ8_0Runs great35 GB6.5
M3 Ultra · 96 GBQ8_0Runs great35 GB12
M3 Ultra · 256 GBQ8_0Runs great35 GB12
M3 Ultra · 512 GBQ8_0Runs great35 GB12
M2 · 8 GBIQ2_XSToo big11 GB
M2 · 16 GBIQ2_STight fit12 GB3.6
M2 · 24 GBQ3_K_SRuns well16 GB2.9
M2 Pro · 16 GBIQ2_STight fit12 GB7
M2 Pro · 32 GBQ4_K_LRuns well21 GB4.5
M2 Max · 32 GBQ4_K_LRuns well21 GB8.6
M2 Max · 64 GBQ6_K_LRuns great28 GB7
M2 Max · 96 GBQ8_0Runs great35 GB6.5
M2 Ultra · 64 GBQ6_K_LRuns great28 GB13
M2 Ultra · 128 GBQ8_0Runs great35 GB12
M2 Ultra · 192 GBQ8_0Runs great35 GB12
M1 · 8 GBIQ2_XSToo big11 GB
M1 · 16 GBIQ2_STight fit12 GB2.4
M1 Pro · 16 GBIQ2_STight fit12 GB6.7
M1 Pro · 32 GBQ4_K_LRuns well21 GB4.4
M1 Max · 32 GBQ4_K_LRuns well21 GB8.3
M1 Max · 64 GBQ6_K_LRuns great28 GB6.7
M1 Ultra · 64 GBQ6_K_LRuns great28 GB12
M1 Ultra · 128 GBQ8_0Runs great35 GB12

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

Information

Maker
Cohere
Released
October 2024
Parameters
32B
Longest chat
8K tokens
Reads pictures
No
Licence
cc-by-nc-4.0
Downloads this month
1,685
Likes
24
Sizes
23, 10 to 65 GB
Kind
Maker's release

Badge for the model card

Shows on Aya Expanse 32B'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