EdgeXpert MS-C931
The clustering-friendly build — a QSFP cable in the box on Gen5 SKUs, sold through both retail and industrial channels.
Price
from ~£3,949 inc VAT
idealo best UK price · Ballicom £4,723.33 (4 TB)
verified 2026-07 · supply & lease options in every proposal
MSI- 128 GB
- Unified memory
- 273 GB/s
- Memory bandwidth
- 1 PFLOP
- Sparse FP4 compute
- 240 W
- Wall power
one coherent CPU+GPU pool
the tok/s ceiling
1,000 AI TOPS
standard socket, near-silent
MSI's EdgeXpert comes in three 4 TB SKUs (Gen4 and Gen5 storage) and is the only family member that ships a QSFP cable in the box on its Gen5 variants — a small thing that says a lot about who it is for: people linking two units on day one.
It is sold through MSI's industrial PC channel as well as mainstream retail, with UK availability at Scan, Novatech and others from ~£3,949.
vs its siblings: QSFP clustering cable included (Gen5 SKUs); industrial/IPC channel availability.
Memory, to scale
128 GB model-visible · bandwidth is the speed limit
Unified LPDDR5x
128 GB · 273 GB/s
LPDDR5x
For scale
DGX Spark — 128 GB @ 273 GB/s
RTX PRO 6000 — 96 GB @ 1.79 TB/s
Mac Studio M3 Ultra — 512 GB @ 819 GB/s
DGX Station GB300 — 748 GB coherent
Two units link directly over ConnectX-7 to serve ~405B-parameter models.
What it actually runs
Declared from research and benchmarks, not computed marketing — tokens-per-second figures are cited where a real measurement exists.
- gpt-oss-120bMXFP4 nativewith headroom~35–40 tok/s (llama.cpp, verified)
- Qwen3.5 122BQ4with headroom10B active — the quality sweet spot at this size
- Nemotron 3 SuperNVFP4with headroomNVIDIA's own stack, 1M context
- DeepSeek V4 FlashINT4 (community)fits~95 GB weights — thin context headroom
- Llama-70B-class denseQ4fitsfits, but dense models are bandwidth-starved here
- 405B-classFP42+ units linkedtwo Sparks linked over ConnectX-7
The full sheet
Compute
- Superchip
- NVIDIA GB10 Grace Blackwell
- GPU
- Blackwell architecture — 6,144 CUDA cores, 5th-gen Tensor Cores
- AI performance
- Up to 1 PFLOP (1,000 TOPS) sparse FP4
- CPU
- 20-core Arm — 10× Cortex-X925 + 10× Cortex-A725
- Interconnect
- NVLink-C2C (CPU ↔ GPU, coherent)
Memory
- Unified memory
- 128 GB LPDDR5x, 256-bit
- Bandwidth
- 273 GB/s
- Model capacity
- ~200B params FP4 inference · 70B fine-tuning · 405B with two linked units
Storage
- NVMe
- 4 TB NVMe (Gen4 or Gen5 by SKU)
Networking & I/O
- Smart NIC
- ConnectX-7, 200 Gb/s total (2× QSFP) — direct 2-unit clustering
- Ethernet / wireless
- 1× 10 GbE RJ-45 · WiFi 7 · BT 5.4
- Ports
- 4× USB4 Type-C · HDMI 2.1a
- Notable
- QSFP cable included in box (12SUS/13SUS SKUs)
Software
- OS
- NVIDIA DGX OS 7 (Ubuntu 24.04 base)
- Stack
- NVIDIA AI stack, NGC catalogue, NIM microservices, Blueprints, DGX Spark playbooks
- Licence
- 90-day NVIDIA AI Enterprise included (3-year purchasable)
Where it shines
- Whole 120B-class models resident in one coherent pool
- Full NVIDIA CUDA stack — the same software as the data centre
- Plugs into a normal socket; 1-litre class, near-silent
- Two-unit clustering built in via ConnectX-7
The trade-offs
- 273 GB/s bandwidth caps single-stream speed (~35–40 tok/s on gpt-oss-120b)
- Not a high-concurrency server — one team, not one company
- Arm (aarch64) platform: x86 containers need rebuilding
- Family took a price rise in Feb 2026 — the RAM shortage is live
Buy this box for
Understanding NVIDIA DGX Spark
GB10 Grace Blackwell Superchip
The DGX Spark is NVIDIA's reference design for desk-scale AI, built on the GB10 Grace Blackwell Superchip: a 20-core Arm CPU and a Blackwell GPU sharing 128 GB of unified LPDDR5x memory over NVLink-C2C. Because CPU and GPU see one memory pool, sparse models up to ~200 billion parameters (FP4) load whole — no VRAM juggling — and it fine-tunes up to 70B.
It ships as the NVIDIA Founders Edition and as near-identical partner machines from ASUS, Dell, HP, MSI, Lenovo, Acer and Gigabyte — the same superchip with different storage, support and price. Two units link directly over their ConnectX-7 200 Gb ports to serve ~405B-parameter models. Every box arrives with DGX OS, the NVIDIA AI stack, NIM microservices and a 90-day NVIDIA AI Enterprise licence.
The trade-off is bandwidth: 273 GB/s of LPDDR5x is the tokens-per-second ceiling (gpt-oss-120b runs at ~35–40 tok/s). It is a superb prototyping and small-team inference machine, not a high-concurrency server. Note the family took an official price rise in February 2026 (Founders Edition $3,999 → $4,699) as the memory shortage bit.
Siblings on the same silicon
NVIDIA
DGX Spark Founders Edition
The reference machine: NVIDIA's own champagne-gold 1.2 kg desk supercomputer, 4 TB standard, dual QSFP exposed for clustering.
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
- AI perf
- 1 PFLOP FP4
£4,899 inc VAT
Scan UK, in stock · US MSRP $4,699 (raised from $3,999, Feb 2026)
Lenovo
ThinkStation PGX
The GB10 goes corporate — ThinkStation build quality, Premier support options, and currently the cheapest UK route into the family.
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
- AI perf
- 1 PFLOP FP4
from ~£3,211 inc VAT
1 TB SKU via UK resellers (Ballicom £2,675.58 ex VAT) · 4 TB ~£3,809 inc
Acer
Veriton GN100
4 TB standard at £3,999.99 — the aggressive value pick that held the old price point through the shortage.
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
- AI perf
- 1 PFLOP FP4
£3,999.99 inc VAT
4 TB standard — best £/TB in the family
ASUS
Ascent GX10
The volume workhorse — 1 TB base SKU, TAA-compliant variant for public sector, stackable chassis design.
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
- AI perf
- 1 PFLOP FP4
£4,299 inc VAT
1 TB SKU, Scan UK in stock · US $3,099.99 (1 TB)
Gigabyte
AI TOP ATOM
The GB10 wired into Gigabyte's AI TOP tuning ecosystem — a neat 1-litre box with 4 TB Gen5 storage.
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
- AI perf
- 1 PFLOP FP4
£4,858.99 inc VAT
Scan UK, in stock · US $3,999.99 (Newegg)
HP
ZGX Nano AI Station G1n
The Z-workstation take on the GB10 — enterprise support and a family ladder that leads up to the ZGX Fury.
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
- AI perf
- 1 PFLOP FP4
~$4,207 · UK from ~£4,939
US verified (CDW, 4 TB) · UK price indicative only — confirm at quote
Sources & verification
Specifications and prices verified 2026-07 against the sources below. The memory shortage is repricing this market monthly — we re-verify at quote.
Compare against the rest of the catalogue, or have us spec it against your workloads.
Back to catalogueSpec the EdgeXpert MS-C931 for your workload
We'll size it against your real token volumes, quote supply and lease options, and stand up the models with evals that prove they do the job — or tell you honestly that a different box (or the API) wins.