Will it run? / gpt-oss 120B / NVIDIA DGX Spark (128GB)

Can the NVIDIA DGX Spark (128GB) run gpt-oss 120B?

Fast Yes, at about 33 tokens/sec (estimated). Replies stream faster than you can read.

33estimated tokens/sec
66.5 GBmemory needed
108.8 GBavailable to the model
273 GB/smemory bandwidth
ollama run gpt-oss:120b

Check the NVIDIA DGX Spark (128GB) price →

How we got this number

Each generated token reads the active part of the model from memory once: 2.8 GB. The NVIDIA DGX Spark (128GB) moves 273 GB/s, so the ceiling is 98 tokens/sec. Real machines reach a fraction of that ceiling; we use 70%, an assumption until we get measured runs for this kind of machine. Send us your real numbers and we'll replace the estimate.

Faster machines for gpt-oss 120B

Mac Studio M5 Ultra (96GB)120+ tok/s$5,499Check price →
Mac Studio M5 Ultra (256GB)120+ tok/s$9,499Check price →
Ryzen AI Max+ 395 mini PC (128GB)31 tok/sCheck price →

Other models on the NVIDIA DGX Spark (128GB)

Qwen 3 30B (A3B)Fast56 tok/s
Qwen3 Coder 30B (A3B)Fast56 tok/s
Qwen 3-VL 30B (A3B)Fast53 tok/s
Qwen3.5 35B (A3B)Fast50 tok/s
Qwen3.6 35B (A3B)Fast50 tok/s
gpt-oss 20BFast46 tok/s