8-bit Open on Hugging Face
Parameters
32B
Weights (est.)
≈ 34.1 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-4-32b-a3b-thinking-8bit --prompt "Hello"
Details
- Base model
- llm-jp/llm-jp-4-32b-a3b-thinking (quantized)
- Vendor
- llm-jp
- Task
- Text generation
- Quantisation
- 8-bit
- Licence
- apache-2.0
- Downloads
- 121 in the last 30 days · 1.9K all time · 1 likes
- Published
- 2026-04-12 · updated 2026-04-12 (7mo ago)
- Languages
- en, ja
- Tags
- qwen3_moe
Other MLX builds of llm-jp/llm-jp-4-32b-a3b-thinking
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| llm-jp-4-32b-a3b-thinking-4bit | 4-bit | ≈ 18.1 GB | 24 GB+ | 121 |
| llm-jp-4-32b-a3b-thinking-8bit (this one) | 8-bit | ≈ 34.1 GB | 48 GB+ | 121 |