Parameters
3.2B
Weights (est.)
≈ 6.4 GB
Mac memory (est.)
16 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/Llama-3.2-3B-Instruct --prompt "Hello"
Details
- Base model
- meta-llama/Llama-3.2-3B-Instruct (finetune) · Llama 3.2 3B Instruct on the radar →
- Vendor
- Meta · Llama
- Task
- Text generation
- Quantisation
- Not stated
- Licence
- llama3.2
- Downloads
- 2.4K in the last 30 days · 1.0M all time · 10 likes
- Published
- 2024-09-25 · updated 2024-09-25 (2y ago)
- Languages
- en, de, fr, it, pt, hi, es, th
- Tags
- llama, facebook, meta, pytorch, llama-3, deploy:azure
Other MLX builds of meta-llama/Llama-3.2-3B-Instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| 3b-es_it-pretrain-research_release-4bit | 4-bit | ≈ 1.9 GB | 8 GB+ | 98 |
| 3b-fr-pretrain-research_release-4bit | 4-bit | ≈ 1.9 GB | 8 GB+ | 97 |
| 3b-zh-pretrain-research_release-4bit | 4-bit | ≈ 1.9 GB | 8 GB+ | 30 |
| 3b-fr-pretrain-research_release-6bit | 6-bit | ≈ 2.7 GB | 8 GB+ | 93 |
| 3b-es_it-pretrain-research_release-6bit | 6-bit | ≈ 2.7 GB | 8 GB+ | 66 |
| 3b-zh-pretrain-research_release-6bit | 6-bit | ≈ 2.7 GB | 8 GB+ | 28 |
| 3b-es_it-pretrain-research_release-8bit | 8-bit | ≈ 3.5 GB | 8 GB+ | 99 |
| 3b-zh-pretrain-research_release-8bit | 8-bit | ≈ 3.5 GB | 8 GB+ | 59 |
| 3b-fr-pretrain-research_release-8bit | 8-bit | ≈ 3.5 GB | 8 GB+ | 54 |
| Llama-3.2-3B-Instruct (this one) | — | ≈ 6.4 GB | 16 GB+ | 2.4K |
| 3b-es_it-pretrain-research_release-bf16 | bf16 | — | — | 81 |
| 3b-zh-pretrain-research_release-bf16 | bf16 | — | — | 63 |
| 3b-fr-pretrain-research_release-bf16 | bf16 | — | — | 47 |