4-bit Open on Hugging Face
Parameters
2.5B
Weights (est.)
≈ 1.4 GB
Mac memory (est.)
8 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/MiniCPM5-2B-chat-mlx-4bit --prompt "Hello"
Details
- Base model
- openbmb/MiniCPM5-2B (quantized)
- Vendor
- openbmb
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 1.0K in the last 30 days · 1.0K all time · 2 likes
- Published
- 2026-10-14 · updated 2026-10-14 (28d ago)
- Tags
- llama, apple-silicon, minicpm, minicpm5, long-context, tool-calling, on-device, edge-ai
Other MLX builds of openbmb/MiniCPM5-2B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| MiniCPM5-2B-mlx-4Bit | 4-bit | ≈ 1.4 GB | 8 GB+ | 5.6K |
| MiniCPM5-2B-chat-mlx-4bit (this one) | 4-bit | ≈ 1.4 GB | 8 GB+ | 1.0K |
| MiniCPM5-2B-OptiQ-4bit | 4-bit | ≈ 1.4 GB | 8 GB+ | 877 |
| MiniCPM5-2B-8bit | 8-bit | ≈ 2.7 GB | 8 GB+ | 1.8K |
| MiniCPM5-2B-chat-mlx-16bit | 16-bit | ≈ 5.0 GB | 8 GB+ | 434 |
| MiniCPM5-2B-bf16 | bf16 | ≈ 5.0 GB | 8 GB+ | 147 |