4-bit Open on Hugging Face
Parameters
340M
Weights (est.)
≈ 192 MB
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-350m-base-4bit --prompt "Hello"
Details
- Base model
- ibm-granite/granite-4.0-h-350m-base (quantized)
- Vendor
- IBM · Granite
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 27 in the last 30 days · 119 all time · 0 likes
- Published
- 2025-10-28 · updated 2025-10-28 (1y ago)
- Tags
- granitemoehybrid, language, granite-4.0
Other MLX builds of ibm-granite/granite-4.0-h-350m-base
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| granite-4.0-h-350m-base-3bit | 3-bit | ≈ 149 MB | 8 GB+ | 24 |
| granite-4.0-h-350m-base-4bit (this one) | 4-bit | ≈ 192 MB | 8 GB+ | 27 |
| granite-4.0-h-350m-base-5bit | 5-bit | ≈ 234 MB | 8 GB+ | 21 |
| granite-4.0-h-350m-base-6bit | 6-bit | ≈ 277 MB | 8 GB+ | 21 |
| granite-4.0-h-350m-base-8bit | 8-bit | ≈ 362 MB | 8 GB+ | 20 |
| granite-4.0-h-350m-base-bf16 | bf16 | ≈ 681 MB | 8 GB+ | 16 |