3-bit Open on Hugging Face
Parameters
6.9B
Weights (est.)
≈ 3.0 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/granite-4.0-h-tiny-3bit-MLX --prompt "Hello"
Details
- Base model
- ibm-granite/granite-4.0-h-tiny (quantized)
- Vendor
- IBM · Granite
- Task
- Text generation
- Quantisation
- 3-bit
- Licence
- apache-2.0
- Downloads
- 57 in the last 30 days · 1.3K all time · 2 likes
- Published
- 2025-10-03 · updated 2025-10-03 (1y ago)
- Languages
- en
- Tags
- granitemoehybrid, granite, apple-silicon, mamba2, transformer, hybrid, long-context, instruct, quantized, MoE
Other MLX builds of ibm-granite/granite-4.0-h-tiny
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| granite-4.0-h-tiny-3bit-MLX (this one) | 3-bit | ≈ 3.0 GB | 8 GB+ | 57 |
| Granite-4.0-H-Tiny-4bit-DWQ | 4-bit DWQ | ≈ 3.9 GB | 8 GB+ | 142 |
| granite-4.0-h-tiny-5bit-MLX | 5-bit | ≈ 4.8 GB | 8 GB+ | 53 |
| granite-4.0-h-tiny-6bit-MLX | 6-bit | ≈ 5.6 GB | 16 GB+ | 40 |
| granite-4.0-h-tiny-8bit | 8-bit | ≈ 7.4 GB | 16 GB+ | 42 |