Outmake Models
Models

Olmo 3.1 32B Think

Ai2 · December 2025

Get Hugging Face

A 32B chat model from Ai2.

Rating
2.9
Downloads
4.6K
Likes
6
Parameters
32B
Context
64K
Licence
apache-2.0

Runs well on M4 · 16 GB

The IQ2_XS size from bartowski is the one to get: a 9.7 GB download that uses 11 GB of memory, and writes about 5.5 words a second.

Ratings

2.9 out of 5, across people's blind votes on its answers.

Text
1286
Coding
1340

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.

Olmo 3.1 32B Think 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 bartowski/allenai_Olmo-3.1-32B-Think-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
IQ2_XXS8.8 GB9.6 GBestimatedOpen
IQ2_XS9.7 GB11 GBestimatedOpen
IQ2_S10 GB11 GBestimatedOpen
IQ2_M11 GB12 GBestimatedOpen
Q2_K12 GB13 GBestimatedOpen
Q2_K_L13 GB13 GBestimatedOpen
IQ3_XXS13 GB13 GBestimatedOpen
IQ3_XS13 GB14 GBestimatedOpen
Q3_K_S14 GB15 GBestimatedOpen
IQ3_M14 GB15 GBestimatedOpen
Q3_K_M16 GB16 GBestimatedOpen
Q3_K_L17 GB18 GBestimatedOpen
IQ4_XS17 GB18 GBestimatedOpen
Q3_K_XL17 GB18 GBestimatedOpen
IQ4_NL18 GB19 GBestimatedOpen
Q4_018 GB19 GBestimatedOpen
Q4_K_S18 GB19 GBestimatedOpen
Q4_K_M19 GB20 GBestimatedOpen
Q4_K_L20 GB21 GBestimatedOpen
Q4_120 GB21 GBestimatedOpen
Q5_K_S22 GB23 GBestimatedOpen
Q5_K_M23 GB24 GBestimatedOpen
Q5_K_L23 GB24 GBestimatedOpen
Q6_K26 GB27 GBestimatedOpen
Q6_K_L27 GB28 GBestimatedOpen
Q8_034 GB35 GBestimatedOpen
BF1664 GB65 GBestimatedOpen

Run It Anywhere

Olmo 3.1 32B Think is an open model, so any app that runs GGUF files can run it. These take the IQ2_XS size from bartowski straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/bartowski/allenai_Olmo-3.1-32B-Think-GGUF:IQ2_XS
  • llama.cppMac, Windows, Linuxllama-server -hf bartowski/allenai_Olmo-3.1-32B-Think-GGUF:IQ2_XS
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/allenai_Olmo-3.1-32B-Think-GGUF:IQ2_XS
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 13 GB
8 GBNone worth runningneeds 13 GB
10 GBNone worth runningneeds 13 GB
11 GBNone worth runningneeds 13 GB
12 GBNone worth runningneeds 13 GB
16 GBIQ3_M14 GB15 GB
20 GBQ4_K_S18 GB19 GB
24 GBQ5_K_M23 GB24 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_XSRuns well11 GB7.9
M6 · 24 GBIQ3_MRuns well16 GB5.7
M6 · 32 GBQ4_1Runs well21 GB4.8
M5 · 16 GBIQ2_XSRuns well11 GB7.2
M5 · 24 GBIQ3_MRuns well16 GB5.2
M5 · 32 GBQ4_1Runs well21 GB4.4
M5 Pro · 24 GBIQ3_MRuns well16 GB10
M5 Pro · 48 GBQ5_K_SRuns great23 GB7
M5 Pro · 64 GBQ6_K_LRuns great30 GB6.2
M5 Max · 36 GBQ5_K_MRuns well24 GB10
M5 Max · 48 GBQ5_K_SRuns great23 GB14
M5 Max · 64 GBQ6_K_LRuns great30 GB12
M5 Max · 128 GBQ8_0Runs great37 GB11
M5 Ultra · 96 GBQ8_0Runs great37 GB21
M5 Ultra · 256 GBQ8_0Runs great37 GB21
M5 Ultra · 512 GBQ8_0Runs great37 GB21
M4 · 16 GBIQ2_XSRuns well11 GB5.5
M4 · 24 GBIQ3_MRuns well16 GB4
M4 · 32 GBQ4_1Runs well21 GB3.5
M4 Pro · 24 GBIQ3_MRuns well16 GB8.5
M4 Pro · 48 GBQ5_K_SRuns great23 GB6
M4 Pro · 64 GBQ6_K_LRuns great30 GB5.2
M4 Max · 36 GBQ5_K_MRuns well24 GB8.5
M4 Max · 48 GBQ5_K_SRuns great23 GB11
M4 Max · 64 GBQ6_K_LRuns great30 GB9.6
M4 Max · 96 GBQ8_0Runs great37 GB6.9
M4 Max · 128 GBQ8_0Runs great37 GB8.9
M3 · 8 GBIQ2_XXSToo big12 GB
M3 · 16 GBIQ2_XSRuns well11 GB4.7
M3 · 24 GBIQ3_MRuns well16 GB3.4
M3 Pro · 18 GBIQ2_MRuns well12 GB6.3
M3 Pro · 36 GBQ5_K_MRuns well24 GB3.4
M3 Max · 36 GBQ5_K_MRuns well24 GB6.5
M3 Max · 48 GBQ5_K_SRuns great23 GB8.7
M3 Max · 64 GBQ6_K_LRuns great30 GB7.5
M3 Max · 96 GBQ8_0Runs great37 GB5.3
M3 Max · 128 GBQ8_0Runs great37 GB6.9
M3 Ultra · 96 GBQ8_0Runs great37 GB13
M3 Ultra · 256 GBQ8_0Runs great37 GB13
M3 Ultra · 512 GBQ8_0Runs great37 GB13
M2 · 8 GBIQ2_XXSToo big12 GB
M2 · 16 GBIQ2_XSRuns well11 GB4.7
M2 · 24 GBIQ3_MRuns well16 GB3.4
M2 Pro · 16 GBIQ2_XSRuns well11 GB8.9
M2 Pro · 32 GBQ4_1Runs well21 GB5.6
M2 Max · 32 GBQ4_1Runs well21 GB11
M2 Max · 64 GBQ6_K_LRuns great30 GB7.5
M2 Max · 96 GBQ8_0Runs great37 GB6.9
M2 Ultra · 64 GBQ6_K_LRuns great30 GB14
M2 Ultra · 128 GBQ8_0Runs great37 GB13
M2 Ultra · 192 GBQ8_0Runs great37 GB13
M1 · 8 GBIQ2_XXSToo big12 GB
M1 · 16 GBIQ2_XSRuns well11 GB3.2
M1 Pro · 16 GBIQ2_XSRuns well11 GB8.3
M1 Pro · 32 GBQ4_1Runs well21 GB5.4
M1 Max · 32 GBQ4_1Runs well21 GB9.8
M1 Max · 64 GBQ6_K_LRuns great30 GB7
M1 Ultra · 64 GBQ6_K_LRuns great30 GB12
M1 Ultra · 128 GBQ8_0Runs great37 GB12

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

Information

Maker
Ai2
Released
December 2025
Parameters
32B
Longest chat
64K tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
4,584
Likes
6
Sizes
27, 8.8 to 64 GB
Kind
Maker's release

Badge for the model card

Shows on Olmo 3.1 32B Think'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