← CanIRunLocal · All hardware · All models
Yes. Llama 3.2 1B fits in fast memory on DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte) at Q8_0 (2 GB needed, 54.4 GB usable), about 158 tok/s (estimate).
Hardware: DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte) · Model: Llama 3.2 1B
DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte) has 64 GB of unified memory and 273 GB/s of memory bandwidth. DGX OS keeps about 15% of unified memory, so Ollama can use 54.4 GB. Llama 3.2 1B (1.24B, Meta) at Q8_0 is a 1.3 GB download and needs 2 GB with headroom for about 8k tokens.
| Best quant | Q8_0 |
|---|---|
| Download size | 1.3 GB |
| Memory needed | 2 GB |
| Usable fast memory | 54.4 GB |
| Fit | fits in fast memory |
| Estimated speed | ~158 tok/s (estimate) |
| Command | ollama run llama3.2:1b |
Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-23.
A project by OnDevice Labs · llms.txt