4-bit Open on Hugging Face
Parameters
73B
Weights (est.)
≈ 41.2 GB
Mac memory (est.)
48 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/llm-jp-3-8x13b-instruct3-4bit --prompt "Hello"
Details
- Base model
- llm-jp/llm-jp-3-8x13b-instruct3 (quantized)
- Vendor
- llm-jp
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 45 in the last 30 days · 367 all time · 1 likes
- Published
- 2025-03-27 · updated 2025-03-27 (1y ago)
- Languages
- en, ja
- Tags
- mixtral
Other MLX builds of llm-jp/llm-jp-3-8x13b-instruct3
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| llm-jp-3-8x13b-instruct3-4bit (this one) | 4-bit | ≈ 41.2 GB | 48 GB+ | 45 |
| llm-jp-3-8x13b-instruct3-8bit | 8-bit | ≈ 77.7 GB | 96 GB+ | 37 |
| llm-jp-3-8x13b-instruct3 | — | ≈ 146 GB | 192 GB+ | 73 |