Different platforms, same budget
NVIDIA DGX Spark vs MINISFORUM MS-S1 MAX
Memory, bandwidth, software stack, price and the models each one can hold. Generated from the two sourced machine records; estimates are labelled.
NVIDIA GB10
NVIDIA DGX Spark
128 GB unified
Open the record
AMD Ryzen AI Max
MINISFORUM MS-S1 MAX
128 GB shared system memory
Open the recordWhat actually differs
- Different software stacks: the NVIDIA DGX Spark runs CUDA, the MINISFORUM MS-S1 MAX runs ROCm and Vulkan. Check that your runtime and model format are supported before comparing numbers.
- Both expose 128 GB to models (128 GB unified vs 128 GB shared system memory).
- Memory bandwidth is not in the record for the MINISFORUM MS-S1 MAX, so the Lab cannot compare generation speed ceilings.
- The MINISFORUM MS-S1 MAX lists about $900 lower (≈ $4.7k vs ≈ $3.8k). Offers are dated snapshots in different regions and tax conditions.
- Form factor: Desktop AI system vs Mini / optional 2U rack.
Which models fit
Estimate · 4-bit · 8K context · 1 request| Model | NVIDIA DGX Spark | MINISFORUM MS-S1 MAX |
|---|---|---|
| Llama 3.1 8B | Fits 16 GB≤ 48 tok/s | Fits 16 GBBandwidth not in record |
| Qwen3 14B | Fits 20 GB≤ 28 tok/s | Fits 20 GBBandwidth not in record |
| Qwen3 32B | Fits 32 GB≤ 13 tok/s | Fits 32 GBBandwidth not in record |
| Llama 3.3 70B | Fits 54 GB≤ 6.3 tok/s | Fits 54 GBBandwidth not in record |
| gpt-oss-20b (MoE) | Fits 23 GB≤ 110 tok/s | Fits 23 GBBandwidth not in record |
| gpt-oss-120b (MoE) | Fits 80 GB≤ 77 tok/s | Fits 80 GBBandwidth not in record |
| Qwen3 235B-A22B (MoE) | Too large 151 GB | Too large 151 GB |
| DeepSeek V3 / R1 671B (MoE, MLA) | Too large 408 GB | Too large 408 GB |
Specifications
Sourced and dated| NVIDIA DGX Spark | MINISFORUM MS-S1 MAX | |
|---|---|---|
| Chip / accelerator | GB10 Grace Blackwell Superchip | Ryzen AI Max+ 395 · Radeon 8060S |
| Platform | NVIDIA GB10 | AMD Ryzen AI Max |
| Model memory | 128 GB unified | 128 GB shared system memory |
| Memory bandwidth | 273 GB/s | Confirm selected configuration |
| Software stack | CUDA | ROCm and Vulkan |
| Operating system | NVIDIA DGX OS | Confirm OS option |
| Storage | 4 TB NVMe | 2 TB SSD |
| Networking | 10 GbE · ConnectX-7 · Wi-Fi 7 | Dual 10 GbE · USB4 v2 · Wi-Fi 7 |
| Form factor | Desktop AI system | Mini / optional 2U rack |
| Dimensions | 150 × 150 × 50.5 mm | Confirm selected configuration |
| Weight | 1.2 kg | Not verified |
| Power rating (not consumption) | 240 W adapter · 140 W SoC TDP | 320 W built-in PSU · processor 130 W sustained / 160 W peak |
| Supplier list price | ≈ $4.7k · Supplier list price · United States · Sept 2026 | ≈ $3.8k · Supplier list price · United States · Sept 2026 |
| Availability | Listed for order · recheck with supplier | Estimated shipping mid-September · supplier statement |
| Model configurations in the Lab | 6 | 2 |
| Setups in the Lab | 5 | 1 |
| Record reviewed | 2026-09-13 | 2026-09-07 |
Before you choose
NVIDIA DGX Spark
- Memory capacity does not establish interactive speed or simultaneous request capacity.
- UT has documented a Spark deployment. Confirm exact software versions and support scope.
MINISFORUM MS-S1 MAX
- The selected 128 GB offer lists estimated mid-September shipping. Confirm delivery before ordering.
- PCIe x16 physical slot is wired PCIe 4.0 x4; do not assume x16 bandwidth. Shared memory allocation depends on runtime and OS.
NVIDIA DGX Spark compared with
ASUS Ascent GX10 Dell Pro Max with GB10 HP ZGX Nano G1n Lenovo ThinkStation PGX Acer Veriton GN100 GIGABYTE AI TOP ATOM MSI EdgeXpert Apple Mac Studio · M3 Ultra Apple Mac Studio · M5 Ultra Framework Desktop · Ryzen AI Max+ HP Z2 Mini G1a AMD Ryzen AI Halo NVIDIA DGX StationMINISFORUM MS-S1 MAX compared with
Framework Desktop · Ryzen AI Max+Own one of these? A measured run (model revision, runtime, context, tokens per second) is worth more than this whole page. Publish it.