Outmake Models
Models

Gemma 2 2B

Google · July 2024

Get Hugging Face

A 2.6B chat model from Google.

Rating
2.3
Downloads
103K
Likes
153
Parameters
2.6B
Context
8K
Licence
gemma

Runs great on M4 · 16 GB

The Q8_0 size from bartowski is the one to get: a 2.8 GB download that uses 3.3 GB of memory, and writes about 16 words a second.

Ratings

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

Text
1200
Coding
1191

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.

Gemma 2 2B 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 l3utterfly/gemma-2-2b-it-gguf on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q2_K1.2 GB1.7 GBestimatedOpen
Q4_01.6 GB2.1 GBestimatedOpen
Q4_0_4_41.6 GB2.1 GBestimatedOpen
Q4_0_4_81.6 GB2.1 GBestimatedOpen
Q4_0_8_81.6 GB2.1 GBestimatedOpen
Q4_K_M1.7 GB2.2 GBestimatedOpen
Q6_K2.2 GB2.6 GBestimatedOpen
Q8_02.8 GB3.3 GBestimatedOpen
F165.2 GB5.7 GBestimatedOpen

Run It Anywhere

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

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/bartowski/gemma-2-2b-it-GGUF:Q8_0
  • llama.cppMac, Windows, Linuxllama-server -hf bartowski/gemma-2-2b-it-GGUF:Q8_0
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/gemma-2-2b-it-GGUF:Q8_0
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBF165.2 GB5.7 GB
8 GBF165.2 GB5.7 GB
10 GBF165.2 GB5.7 GB
11 GBF165.2 GB5.7 GB
12 GBF165.2 GB5.7 GB
16 GBF165.2 GB5.7 GB
20 GBF165.2 GB5.7 GB
24 GBF165.2 GB5.7 GB
32 GBF165.2 GB5.7 GB
48 GBF165.2 GB5.7 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 great3.3 GB25
M6 · 24 GBQ8_0Runs great3.3 GB25
M6 · 32 GBQ8_0Runs great3.3 GB25
M5 · 16 GBQ8_0Runs great3.3 GB22
M5 · 24 GBQ8_0Runs great3.3 GB22
M5 · 32 GBQ8_0Runs great3.3 GB22
M5 Pro · 24 GBQ8_0Runs great3.3 GB44
M5 Pro · 48 GBQ8_0Runs great3.3 GB44
M5 Pro · 64 GBQ8_0Runs great3.3 GB44
M5 Max · 36 GBQ8_0Runs great3.3 GB64
M5 Max · 48 GBQ8_0Runs great3.3 GB82
M5 Max · 64 GBQ8_0Runs great3.3 GB82
M5 Max · 128 GBQ8_0Runs great3.3 GB82
M5 Ultra · 96 GBQ8_0Runs great3.3 GB148
M5 Ultra · 256 GBQ8_0Runs great3.3 GB148
M5 Ultra · 512 GBQ8_0Runs great3.3 GB148
M4 · 16 GBQ8_0Runs great3.3 GB16
M4 · 24 GBQ8_0Runs great3.3 GB16
M4 · 32 GBQ8_0Runs great3.3 GB16
M4 Pro · 24 GBQ8_0Runs great3.3 GB34
M4 Pro · 48 GBQ8_0Runs great3.3 GB34
M4 Pro · 64 GBQ8_0Runs great3.3 GB34
M4 Max · 36 GBQ8_0Runs great3.3 GB46
M4 Max · 48 GBQ8_0Runs great3.3 GB57
M4 Max · 64 GBQ8_0Runs great3.3 GB57
M4 Max · 96 GBQ8_0Runs great3.3 GB46
M4 Max · 128 GBQ8_0Runs great3.3 GB57
M3 · 8 GBQ8_0Runs great3.3 GB14
M3 · 16 GBQ8_0Runs great3.3 GB14
M3 · 24 GBQ8_0Runs great3.3 GB14
M3 Pro · 18 GBQ8_0Runs great3.3 GB20
M3 Pro · 36 GBQ8_0Runs great3.3 GB20
M3 Max · 36 GBQ8_0Runs great3.3 GB37
M3 Max · 48 GBQ8_0Runs great3.3 GB47
M3 Max · 64 GBQ8_0Runs great3.3 GB47
M3 Max · 96 GBQ8_0Runs great3.3 GB37
M3 Max · 128 GBQ8_0Runs great3.3 GB47
M3 Ultra · 96 GBQ8_0Runs great3.3 GB78
M3 Ultra · 256 GBQ8_0Runs great3.3 GB78
M3 Ultra · 512 GBQ8_0Runs great3.3 GB78
M2 · 8 GBQ8_0Runs great3.3 GB14
M2 · 16 GBQ8_0Runs great3.3 GB14
M2 · 24 GBQ8_0Runs great3.3 GB14
M2 Pro · 16 GBQ8_0Runs great3.3 GB26
M2 Pro · 32 GBQ8_0Runs great3.3 GB26
M2 Max · 32 GBQ8_0Runs great3.3 GB47
M2 Max · 64 GBQ8_0Runs great3.3 GB47
M2 Max · 96 GBQ8_0Runs great3.3 GB47
M2 Ultra · 64 GBQ8_0Runs great3.3 GB77
M2 Ultra · 128 GBQ8_0Runs great3.3 GB77
M2 Ultra · 192 GBQ8_0Runs great3.3 GB77
M1 · 8 GBQ8_0Runs great3.3 GB9.5
M1 · 16 GBQ8_0Runs great3.3 GB9.5
M1 Pro · 16 GBQ8_0Runs great3.3 GB24
M1 Pro · 32 GBQ8_0Runs great3.3 GB24
M1 Max · 32 GBQ8_0Runs great3.3 GB41
M1 Max · 64 GBQ8_0Runs great3.3 GB41
M1 Ultra · 64 GBQ8_0Runs great3.3 GB62
M1 Ultra · 128 GBQ8_0Runs great3.3 GB62

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
Google
Released
July 2024
Parameters
2.6B
Longest chat
8K tokens
Reads pictures
No
Licence
gemma
Downloads this month
102,757
Likes
153
Sizes
9, 1.2 to 5.2 GB
Kind
Maker's release

Badge for the model card

Shows on Gemma 2 2B'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