Outmake Models
Models

Gemma 3n E4B

Google · June 2025

Get Hugging Face

A 6.9B chat model from Google that reads pictures as well as text.

Rating
3.2
Downloads
51K
Likes
229
Parameters
6.9B
Context
32K
Licence
gemma

Runs great on M4 · 16 GB

The Q8_0 size from unsloth is the one to get: a 7.4 GB download that uses 7.6 GB of memory, and writes about 9.7 words a second.

Ratings

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

Text
1318
Coding
1307

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 3n E4B 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 unsloth/gemma-3n-E4B-it-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
UD-IQ2_XXS2.8 GB3.0 GBestimatedOpen
UD-IQ2_M3.1 GB3.3 GBestimatedOpen
Q2_K3.2 GB3.4 GBestimatedOpen
Q2_K_L3.2 GB3.4 GBestimatedOpen
UD-IQ3_XXS3.3 GB3.5 GBestimatedOpen
Q3_K_S3.5 GB3.7 GBestimatedOpen
UD-Q2_K_XL3.7 GB3.9 GBestimatedOpen
Q3_K_M3.7 GB3.9 GBestimatedOpen
UD-Q3_K_XL4.1 GB4.3 GBestimatedOpen
IQ4_XS4.3 GB4.5 GBestimatedOpen
IQ4_NL4.4 GB4.6 GBestimatedOpen
Q4_04.4 GB4.6 GBestimatedOpen
Q4_K_S4.4 GB4.6 GBestimatedOpen
Q4_K_M4.5 GB4.7 GBestimatedOpen
Q4_14.6 GB4.8 GBestimatedOpen
Q5_K_S4.9 GB5.1 GBestimatedOpen
Q5_K_M5.0 GB5.2 GBestimatedOpen
UD-Q4_K_XL5.4 GB5.6 GBestimatedOpen
UD-Q5_K_XL5.9 GB6.0 GBestimatedOpen
Q6_K6.3 GB6.5 GBestimatedOpen
UD-Q6_K_XL6.6 GB6.8 GBestimatedOpen
Q8_07.4 GB7.5 GBestimatedOpen
UD-Q8_K_XL11 GB11 GBestimatedOpen
F1614 GB14 GBestimatedOpen

Run It Anywhere

Gemma 3n E4B is an open model, so any app that runs GGUF files can run it. These take the Q8_0 size from unsloth straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/unsloth/gemma-3n-E4B-it-GGUF:Q8_0
  • llama.cppMac, Windows, Linuxllama-server -hf unsloth/gemma-3n-E4B-it-GGUF:Q8_0
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/unsloth/gemma-3n-E4B-it-GGUF:Q8_0
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBUD-Q4_K_XL5.4 GB5.6 GB
8 GBQ8_07.4 GB7.5 GB
10 GBQ8_07.4 GB7.5 GB
11 GBUD-Q8_K_XL11 GB11 GB
12 GBUD-Q8_K_XL11 GB11 GB
16 GBF1614 GB14 GB
20 GBF1614 GB14 GB
24 GBF1614 GB14 GB
32 GBF1614 GB14 GB
48 GBF1614 GB14 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 great7.6 GB14
M6 · 24 GBQ8_0Runs great7.7 GB14
M6 · 32 GBQ8_0Runs great7.7 GB14
M5 · 16 GBQ8_0Runs great7.6 GB13
M5 · 24 GBQ8_0Runs great7.7 GB13
M5 · 32 GBQ8_0Runs great7.7 GB13
M5 Pro · 24 GBQ8_0Runs great7.7 GB26
M5 Pro · 48 GBQ8_0Runs great7.7 GB26
M5 Pro · 64 GBQ8_0Runs great7.7 GB26
M5 Max · 36 GBQ8_0Runs great7.7 GB37
M5 Max · 48 GBQ8_0Runs great7.7 GB49
M5 Max · 64 GBQ8_0Runs great7.7 GB49
M5 Max · 128 GBQ8_0Runs great7.7 GB49
M5 Ultra · 96 GBQ8_0Runs great7.7 GB89
M5 Ultra · 256 GBQ8_0Runs great7.7 GB89
M5 Ultra · 512 GBQ8_0Runs great7.7 GB89
M4 · 16 GBQ8_0Runs great7.6 GB9.7
M4 · 24 GBQ8_0Runs great7.7 GB9.7
M4 · 32 GBQ8_0Runs great7.7 GB9.7
M4 Pro · 24 GBQ8_0Runs great7.7 GB20
M4 Pro · 48 GBQ8_0Runs great7.7 GB20
M4 Pro · 64 GBQ8_0Runs great7.7 GB20
M4 Max · 36 GBQ8_0Runs great7.7 GB28
M4 Max · 48 GBQ8_0Runs great7.7 GB35
M4 Max · 64 GBQ8_0Runs great7.7 GB35
M4 Max · 96 GBQ8_0Runs great7.7 GB28
M4 Max · 128 GBQ8_0Runs great7.7 GB35
M3 · 8 GBUD-Q2_K_XLRuns great3.9 GB13
M3 · 16 GBQ8_0Runs great7.6 GB8.2
M3 · 24 GBQ8_0Runs great7.7 GB8.2
M3 Pro · 18 GBQ8_0Runs great7.7 GB12
M3 Pro · 36 GBQ8_0Runs great7.7 GB12
M3 Max · 36 GBQ8_0Runs great7.7 GB22
M3 Max · 48 GBQ8_0Runs great7.7 GB29
M3 Max · 64 GBQ8_0Runs great7.7 GB29
M3 Max · 96 GBQ8_0Runs great7.7 GB22
M3 Max · 128 GBQ8_0Runs great7.7 GB29
M3 Ultra · 96 GBQ8_0Runs great7.7 GB49
M3 Ultra · 256 GBQ8_0Runs great7.7 GB49
M3 Ultra · 512 GBQ8_0Runs great7.7 GB49
M2 · 8 GBUD-Q2_K_XLRuns great3.9 GB13
M2 · 16 GBQ8_0Runs great7.6 GB8.2
M2 · 24 GBQ8_0Runs great7.7 GB8.2
M2 Pro · 16 GBQ8_0Runs great7.6 GB16
M2 Pro · 32 GBQ8_0Runs great7.7 GB16
M2 Max · 32 GBQ8_0Runs great7.7 GB28
M2 Max · 64 GBQ8_0Runs great7.7 GB28
M2 Max · 96 GBQ8_0Runs great7.7 GB28
M2 Ultra · 64 GBQ8_0Runs great7.7 GB48
M2 Ultra · 128 GBQ8_0Runs great7.7 GB48
M2 Ultra · 192 GBQ8_0Runs great7.7 GB48
M1 · 8 GBUD-Q2_K_XLRuns great3.9 GB8.8
M1 · 16 GBQ8_0Runs great7.6 GB5.5
M1 Pro · 16 GBQ8_0Runs great7.6 GB15
M1 Pro · 32 GBQ8_0Runs great7.7 GB15
M1 Max · 32 GBQ8_0Runs great7.7 GB25
M1 Max · 64 GBQ8_0Runs great7.7 GB25
M1 Ultra · 64 GBQ8_0Runs great7.7 GB40
M1 Ultra · 128 GBQ8_0Runs great7.7 GB40

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

Packaged by

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

Information

Maker
Google
Released
June 2025
Parameters
6.9B
Longest chat
32K tokens
Reads pictures
Yes
Licence
gemma
Downloads this month
51,353
Likes
229
Sizes
24, 2.8 to 14 GB
Kind
Maker's release

Badge for the model card

Shows on Gemma 3n E4B'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