Outmake Models
Models

Bonsai 1.7B

PrismML · March 2026

Get Hugging Face

A 1.7B chat model from PrismML.

Downloads
132K
Likes
97
Parameters
1.7B
Context
32K
Licence
apache-2.0

Runs great on M4 · 16 GB

The Q1_0 size from prism-ml is the one to get: a 0.2 GB download that uses 2.5 GB of memory, and writes about 18 words a second.

Bonsai 1.7B 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 prism-ml/Bonsai-1.7B-gguf on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q1_00.2 GB0.9 GBestimatedOpen
Bonsai-1.7B0.2 GB0.9 GBestimatedOpen

Run It Anywhere

Bonsai 1.7B is an open model, so any app that runs GGUF files can run it. These take the Q1_0 size from prism-ml straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/prism-ml/Bonsai-1.7B-gguf:Q1_0
  • llama.cppMac, Windows, Linuxllama-server -hf prism-ml/Bonsai-1.7B-gguf:Q1_0
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/prism-ml/Bonsai-1.7B-gguf:Q1_0
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBBonsai-1.7B0.2 GB0.9 GB
8 GBBonsai-1.7B0.2 GB0.9 GB
10 GBBonsai-1.7B0.2 GB0.9 GB
11 GBBonsai-1.7B0.2 GB0.9 GB
12 GBBonsai-1.7B0.2 GB0.9 GB
16 GBBonsai-1.7B0.2 GB0.9 GB
20 GBBonsai-1.7B0.2 GB0.9 GB
24 GBBonsai-1.7B0.2 GB0.9 GB
32 GBBonsai-1.7B0.2 GB0.9 GB
48 GBBonsai-1.7B0.2 GB0.9 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 GBQ1_0Runs great2.5 GB28
M6 · 24 GBQ1_0Runs great2.5 GB28
M6 · 32 GBQ1_0Runs great2.5 GB28
M5 · 16 GBQ1_0Runs great2.5 GB25
M5 · 24 GBQ1_0Runs great2.5 GB25
M5 · 32 GBQ1_0Runs great2.5 GB25
M5 Pro · 24 GBQ1_0Runs great2.5 GB49
M5 Pro · 48 GBQ1_0Runs great2.5 GB49
M5 Pro · 64 GBQ1_0Runs great2.5 GB49
M5 Max · 36 GBQ1_0Runs great2.5 GB71
M5 Max · 48 GBQ1_0Runs great2.5 GB91
M5 Max · 64 GBQ1_0Runs great2.5 GB91
M5 Max · 128 GBQ1_0Runs great2.5 GB91
M5 Ultra · 96 GBQ1_0Runs great2.5 GB161
M5 Ultra · 256 GBQ1_0Runs great2.5 GB161
M5 Ultra · 512 GBQ1_0Runs great2.5 GB161
M4 · 16 GBQ1_0Runs great2.5 GB18
M4 · 24 GBQ1_0Runs great2.5 GB18
M4 · 32 GBQ1_0Runs great2.5 GB18
M4 Pro · 24 GBQ1_0Runs great2.5 GB37
M4 Pro · 48 GBQ1_0Runs great2.5 GB37
M4 Pro · 64 GBQ1_0Runs great2.5 GB37
M4 Max · 36 GBQ1_0Runs great2.5 GB50
M4 Max · 48 GBQ1_0Runs great2.5 GB61
M4 Max · 64 GBQ1_0Runs great2.5 GB61
M4 Max · 96 GBQ1_0Runs great2.5 GB50
M4 Max · 128 GBQ1_0Runs great2.5 GB61
M3 · 8 GBQ1_0Runs great2.5 GB16
M3 · 16 GBQ1_0Runs great2.5 GB16
M3 · 24 GBQ1_0Runs great2.5 GB16
M3 Pro · 18 GBQ1_0Runs great2.5 GB23
M3 Pro · 36 GBQ1_0Runs great2.5 GB23
M3 Max · 36 GBQ1_0Runs great2.5 GB41
M3 Max · 48 GBQ1_0Runs great2.5 GB51
M3 Max · 64 GBQ1_0Runs great2.5 GB51
M3 Max · 96 GBQ1_0Runs great2.5 GB41
M3 Max · 128 GBQ1_0Runs great2.5 GB51
M3 Ultra · 96 GBQ1_0Runs great2.5 GB82
M3 Ultra · 256 GBQ1_0Runs great2.5 GB82
M3 Ultra · 512 GBQ1_0Runs great2.5 GB82
M2 · 8 GBQ1_0Runs great2.5 GB16
M2 · 16 GBQ1_0Runs great2.5 GB16
M2 · 24 GBQ1_0Runs great2.5 GB16
M2 Pro · 16 GBQ1_0Runs great2.5 GB29
M2 Pro · 32 GBQ1_0Runs great2.5 GB29
M2 Max · 32 GBQ1_0Runs great2.5 GB51
M2 Max · 64 GBQ1_0Runs great2.5 GB51
M2 Max · 96 GBQ1_0Runs great2.5 GB51
M2 Ultra · 64 GBQ1_0Runs great2.5 GB81
M2 Ultra · 128 GBQ1_0Runs great2.5 GB81
M2 Ultra · 192 GBQ1_0Runs great2.5 GB81
M1 · 8 GBQ1_0Runs great2.5 GB11
M1 · 16 GBQ1_0Runs great2.5 GB11
M1 Pro · 16 GBQ1_0Runs great2.5 GB27
M1 Pro · 32 GBQ1_0Runs great2.5 GB27
M1 Max · 32 GBQ1_0Runs great2.5 GB44
M1 Max · 64 GBQ1_0Runs great2.5 GB44
M1 Ultra · 64 GBQ1_0Runs great2.5 GB64
M1 Ultra · 128 GBQ1_0Runs great2.5 GB64

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

Information

Maker
PrismML
Released
March 2026
Parameters
1.7B
Longest chat
32K tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
131,612
Likes
97
Sizes
2, 0.2 to 0.2 GB
Kind
Maker's release

Badge for the model card

Shows on Bonsai 1.7B'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