3-bit Open on Hugging Face
Parameters
3.2B
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/Falcon3-3B-Instruct-3bit --prompt "Hello"
Details
- Base model
- tiiuae/Falcon3-3B-Instruct (quantized)
- Vendor
- TII · Falcon
- Task
- Text generation
- Quantisation
- 3-bit
- Licence
- falcon-llm-license
- Downloads
- 17 in the last 30 days · 387 all time · 0 likes
- Published
- 2024-12-17 · updated 2024-12-17 (1y ago)
- Languages
- en, fr, es, pt
- Tags
- llama, falcon3
Other MLX builds of tiiuae/Falcon3-3B-Instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Falcon3-3B-Instruct-3bit (this one) | 3-bit | ≈ 1.4 GB | 8 GB+ | 17 |
| Falcon3-3B-Instruct-4bit | 4-bit | ≈ 1.8 GB | 8 GB+ | 122 |
| Falcon3-3B-Instruct-6bit | 6-bit | ≈ 2.6 GB | 8 GB+ | 36 |
| Falcon3-3B-Instruct-8bit | 8-bit | ≈ 3.4 GB | 8 GB+ | 29 |
| Falcon3-3B-Instruct-bf16 | bf16 | ≈ 6.5 GB | 16 GB+ | 28 |