Hugging FaceA 8.2B chat model from PrismML.
- Downloads
- 93K
- Likes
- 800
- Parameters
- 8.2B
- Context
- 64K
- Licence
- apache-2.0
Runs great on M4 · 16 GB
The Q1_0 size from prism-ml is the one to get: a 1.2 GB download that uses 6.9 GB of memory, and writes about 12 words a second.
Bonsai 8B 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-8B-gguf on Hugging Face. A smaller size is a smaller download and a little less accurate.
Run It Anywhere
Bonsai 8B 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, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/prism-ml/Bonsai-8B-gguf:Q1_0 - llama.cppMac, Windows, Linux
llama-server -hf prism-ml/Bonsai-8B-gguf:Q1_0 - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/prism-ml/Bonsai-8B-gguf:Q1_0
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 8 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 10 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 11 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 12 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 16 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 20 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 24 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 32 GB | Bonsai-8B | 1.2 GB | 2.0 GB |
| 48 GB | Bonsai-8B | 1.2 GB | 2.0 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
| Mac | Size | Fit | Memory | Writes |
|---|---|---|---|---|
| M6 · 16 GB | Q1_0 | Runs great | 6.9 GB | 18 |
| M6 · 24 GB | Q1_0 | Runs great | 6.9 GB | 18 |
| M6 · 32 GB | Q1_0 | Runs great | 6.9 GB | 18 |
| M5 · 16 GB | Q1_0 | Runs great | 6.9 GB | 17 |
| M5 · 24 GB | Q1_0 | Runs great | 6.9 GB | 17 |
| M5 · 32 GB | Q1_0 | Runs great | 6.9 GB | 17 |
| M5 Pro · 24 GB | Q1_0 | Runs great | 6.9 GB | 33 |
| M5 Pro · 48 GB | Q1_0 | Runs great | 6.9 GB | 33 |
| M5 Pro · 64 GB | Q1_0 | Runs great | 6.9 GB | 33 |
| M5 Max · 36 GB | Q1_0 | Runs great | 6.9 GB | 48 |
| M5 Max · 48 GB | Q1_0 | Runs great | 6.9 GB | 62 |
| M5 Max · 64 GB | Q1_0 | Runs great | 6.9 GB | 62 |
| M5 Max · 128 GB | Q1_0 | Runs great | 6.9 GB | 62 |
| M5 Ultra · 96 GB | Q1_0 | Runs great | 6.9 GB | 110 |
| M5 Ultra · 256 GB | Q1_0 | Runs great | 6.9 GB | 110 |
| M5 Ultra · 512 GB | Q1_0 | Runs great | 6.9 GB | 110 |
| M4 · 16 GB | Q1_0 | Runs great | 6.9 GB | 12 |
| M4 · 24 GB | Q1_0 | Runs great | 6.9 GB | 12 |
| M4 · 32 GB | Q1_0 | Runs great | 6.9 GB | 12 |
| M4 Pro · 24 GB | Q1_0 | Runs great | 6.9 GB | 25 |
| M4 Pro · 48 GB | Q1_0 | Runs great | 6.9 GB | 25 |
| M4 Pro · 64 GB | Q1_0 | Runs great | 6.9 GB | 25 |
| M4 Max · 36 GB | Q1_0 | Runs great | 6.9 GB | 35 |
| M4 Max · 48 GB | Q1_0 | Runs great | 6.9 GB | 43 |
| M4 Max · 64 GB | Q1_0 | Runs great | 6.9 GB | 43 |
| M4 Max · 96 GB | Q1_0 | Runs great | 6.9 GB | 35 |
| M4 Max · 128 GB | Q1_0 | Runs great | 6.9 GB | 43 |
| M3 · 8 GB | Q1_0 | Runs great | 2.7 GB | 11 |
| M3 · 16 GB | Q1_0 | Runs great | 6.9 GB | 11 |
| M3 · 24 GB | Q1_0 | Runs great | 6.9 GB | 11 |
| M3 Pro · 18 GB | Q1_0 | Runs great | 6.9 GB | 16 |
| M3 Pro · 36 GB | Q1_0 | Runs great | 6.9 GB | 16 |
| M3 Max · 36 GB | Q1_0 | Runs great | 6.9 GB | 28 |
| M3 Max · 48 GB | Q1_0 | Runs great | 6.9 GB | 36 |
| M3 Max · 64 GB | Q1_0 | Runs great | 6.9 GB | 36 |
| M3 Max · 96 GB | Q1_0 | Runs great | 6.9 GB | 28 |
| M3 Max · 128 GB | Q1_0 | Runs great | 6.9 GB | 36 |
| M3 Ultra · 96 GB | Q1_0 | Runs great | 6.9 GB | 58 |
| M3 Ultra · 256 GB | Q1_0 | Runs great | 6.9 GB | 58 |
| M3 Ultra · 512 GB | Q1_0 | Runs great | 6.9 GB | 58 |
| M2 · 8 GB | Q1_0 | Runs great | 2.7 GB | 11 |
| M2 · 16 GB | Q1_0 | Runs great | 6.9 GB | 11 |
| M2 · 24 GB | Q1_0 | Runs great | 6.9 GB | 11 |
| M2 Pro · 16 GB | Q1_0 | Runs great | 6.9 GB | 20 |
| M2 Pro · 32 GB | Q1_0 | Runs great | 6.9 GB | 20 |
| M2 Max · 32 GB | Q1_0 | Runs great | 6.9 GB | 35 |
| M2 Max · 64 GB | Q1_0 | Runs great | 6.9 GB | 35 |
| M2 Max · 96 GB | Q1_0 | Runs great | 6.9 GB | 35 |
| M2 Ultra · 64 GB | Q1_0 | Runs great | 6.9 GB | 58 |
| M2 Ultra · 128 GB | Q1_0 | Runs great | 6.9 GB | 58 |
| M2 Ultra · 192 GB | Q1_0 | Runs great | 6.9 GB | 58 |
| M1 · 8 GB | Q1_0 | Runs great | 2.7 GB | 7.2 |
| M1 · 16 GB | Q1_0 | Runs great | 6.9 GB | 7.2 |
| M1 Pro · 16 GB | Q1_0 | Runs great | 6.9 GB | 18 |
| M1 Pro · 32 GB | Q1_0 | Runs great | 6.9 GB | 18 |
| M1 Max · 32 GB | Q1_0 | Runs great | 6.9 GB | 31 |
| M1 Max · 64 GB | Q1_0 | Runs great | 6.9 GB | 31 |
| M1 Ultra · 64 GB | Q1_0 | Runs great | 6.9 GB | 46 |
| M1 Ultra · 128 GB | Q1_0 | Runs great | 6.9 GB | 46 |
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
- 8.2B
- Longest chat
- 64K tokens
- Reads pictures
- No
- Licence
- apache-2.0
- Downloads this month
- 93,397
- Likes
- 800
- Sizes
- 2, 1.2 to 1.2 GB
- Kind
- Maker's release
- Source
- prism-ml/Bonsai-8B-gguf
Badge for the model card
Shows on Bonsai 8B's Hugging Face card which Macs run it well, linked to this page. It updates every morning.
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.