estimate, except measured file sizes
Can I run Llama 3.2 3B locally?
3.21B parameters. Parameter count is an estimate.
- Licence:
- not recorded
- Context:
- 131,072 tokens
- Modality:
- not recorded
- Released:
- not recorded
Small enough for a Mac mini or a modest GPU. A good stack to test your setup.
· Architecture source · Model page
Memory per quant
KV is at 8,192 tokens. Overhead is 0.8 GB. Both are estimates. A measured file size is labelled. The last column is the cheapest machine that fits the total fully.
| Quant | File size | KV at 8K | +0.8 overhead | Total | Cheapest full fit |
|---|---|---|---|---|---|
| Q8 | 3.4 GB est. | 0.9 GB est. | 0.8 GB est. | 5.2 GB est. | GeForce RTX 3060 12GB, €257 (used (median of sold listings)) |
| Q4 | 1.9 GB est. | 0.9 GB est. | 0.8 GB est. | 3.7 GB est. | GeForce RTX 3060 12GB, €257 (used (median of sold listings)) |
Cheapest rig that fits
Q4, full: GeForce RTX 3060 12GB, €257 (used (median of sold listings))
Q4, slow, partly in system RAM: Nothing on our list of 29 machines fits this quant.