← CanIRunLocal · All hardware · All models

Can a DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) run Ornith-1.5 397B (MoE)?

No. Ornith-1.5 397B (MoE) does not fit on DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte): Default needs 246.4 GB and 108.8 GB is usable.

Hardware: DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) · Model: Ornith-1.5 397B (MoE)

DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) has 128 GB of unified memory and 273 GB/s of memory bandwidth. DGX OS keeps about 15% of unified memory, so Ollama can use 108.8 GB. Ornith-1.5 397B (MoE) (397B, Ornith) at Default is a 242 GB download and needs 246.4 GB with headroom for about 8k tokens.

Best quantDefault
Download size242 GB
Memory needed246.4 GB
Usable fast memory108.8 GB
Fitdoes not fit
Estimated speedno estimate, it does not fit
Commandollama run ornith-1.5:397b

Smallest hardware that runs it: 4x Mac Studio M5 Max (40c GPU) 128GB at Default (fits in fast memory, 246.4 GB needed, 378.0 GB usable, about 4 tok/s (estimate)).

A smaller Ornith model that fits this hardware: Ornith-1.5 35B-A3B (MoE) at Default (fits in fast memory, 24.8 GB). Can a DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) run Ornith-1.5 35B-A3B (MoE)?

Open DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) in the checker

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