fp16 Open on Hugging Face
Parameters
8.2B
Weights (est.)
≈ 16.3 GB
Mac memory (est.)
24 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-3.3-8b-instruct-fp16 --prompt "Hello"
Details
- Base model
- ibm-granite/granite-3.3-8b-instruct (finetune)
- Vendor
- IBM · Granite
- Task
- Text generation
- Quantisation
- fp16
- Licence
- apache-2.0
- Downloads
- 56 in the last 30 days · 2.4K all time · 3 likes
- Published
- 2025-04-16 · updated 2025-04-16 (1y ago)
- Tags
- granite, language, granite-3.3
Other MLX builds of ibm-granite/granite-3.3-8b-instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| granite-3.3-8b-instruct-4bit | 4-bit | ≈ 4.6 GB | 8 GB+ | 207 |
| granite-3.3-8b-instruct-6bit | 6-bit | ≈ 6.6 GB | 16 GB+ | 27 |
| granite-3.3-8b-instruct-8bit | 8-bit | ≈ 8.7 GB | 16 GB+ | 45 |
| granite-3.3-8b-instruct-fp16 (this one) | fp16 | ≈ 16.3 GB | 24 GB+ | 56 |