3-bit Open on Hugging Face
Parameters
10B
Weights (est.)
≈ 4.5 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/INTELLECT-1-Instruct-3bit --prompt "Hello"
Details
- Base model
- PrimeIntellect/INTELLECT-1-Instruct (quantized)
- Vendor
- PrimeIntellect
- Task
- Text generation
- Quantisation
- 3-bit
- Licence
- apache-2.0
- Downloads
- 61 in the last 30 days · 299 all time · 0 likes
- Published
- 2024-11-29 · updated 2024-11-30 (1y ago)
- Languages
- en
- Tags
- llama
Other MLX builds of PrimeIntellect/INTELLECT-1-Instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| INTELLECT-1-Instruct-3bit (this one) | 3-bit | ≈ 4.5 GB | 8 GB+ | 61 |
| INTELLECT-1-Instruct-4bit | 4-bit | ≈ 5.7 GB | 16 GB+ | 127 |
| INTELLECT-1-Instruct-6bit | 6-bit | ≈ 8.3 GB | 16 GB+ | 32 |
| INTELLECT-1-Instruct-8bit | 8-bit | ≈ 10.8 GB | 16 GB+ | 39 |
| INTELLECT-1-Instruct-bf16 | bf16 | ≈ 20.4 GB | 32 GB+ | 80 |