EVO-X2
The cheap seat — 128 GB of unified memory from ~£1,660, if you can live with consumer-grade support.
Price
£2,099 inc VAT
128 GB/2 TB verified (Amazon UK) · gmktec.uk shows £1,659.99 (tier ambiguous)
verified 2026-07 · supply & lease options in every proposal
GMKtec- 128 GB
- Unified memory
- 256 GB/s
- Memory bandwidth
- 2.8 L
- Chassis
- ~£1.7–2.1k
- UK street
LPDDR5X-8000
the platform ceiling
dual M.2 to 16 TB
cheapest 128 GB path
GMKtec's EVO-X2 is the lowest-cost door into 128 GB unified memory: verified UK retail at £2,099 for the 128 GB/2 TB build (Amazon UK), with GMKtec's own UK store showing £1,659.99 (configuration tier ambiguous — confirm before ordering).
You give up networking (2.5 GbE only), support depth and acoustic refinement versus HP and Framework — but for a lab box proving what a £2k machine can do, nothing touches the price.
vs its siblings: Price leadership, full stop.
Memory, to scale
128 GB model-visible · bandwidth is the speed limit
Unified LPDDR5X
128 GB · 256 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
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-120bMXFP4with headroom30–55 tok/s — same silicon as the dearer boxes
- Qwen3.6 35BQ4with headroom~66 tok/s
- gpt-oss-20bMXFP4with headroompipeline workhorse
- Qwen3-235B-classQ3fits~11 tok/s — party trick, not production
The full sheet
Compute
- APU
- AMD Ryzen AI Max+ 395 — 16× Zen 5, Radeon 8060S, XDNA 2
Memory
- Unified memory
- 128 GB LPDDR5X-8000
Storage & I/O
- Storage
- Dual M.2 (to 16 TB)
- Networking
- 2.5 GbE · Wi-Fi 7 · 230 W brick
Where it shines
- The cheapest 128 GB unified-memory machine in the catalogue
- Same LLM performance as Strix boxes costing £1k more
- Compact and easily redeployed
The trade-offs
- 2.5 GbE only — no serious clustering or fast NAS path
- Consumer-grade warranty and support
- Fan noise under sustained load
- Price/config ambiguity on the vendor store — verify the tier
Buy this box for
Understanding AMD Strix Halo
Ryzen AI Max+ 395
AMD's Ryzen AI Max+ 395 ('Strix Halo') puts 16 Zen 5 cores, a 40-CU Radeon 8060S GPU and up to 128 GB of unified LPDDR5X-8000 on one package. Framework, HP, GMKtec, Beelink, Corsair and AMD itself all ship it as a compact desktop, from roughly £1,700 to £2,700 in the UK.
It is a mixture-of-experts machine: gpt-oss-120b runs at 30–55 tok/s and even Qwen3-235B squeezes in at 3-bit (~11 tok/s), but dense 70B models crawl at ~5 tok/s because 256 GB/s of bandwidth is the ceiling. The software stack (Vulkan / ROCm / llama.cpp / LM Studio) is genuinely usable in 2026 but remains a step behind CUDA for production serving and fine-tuning.
It is the right first box for proving local AI on real workloads before committing to bigger silicon — and the refresh is close: AMD's Ryzen AI Max 400 series ('Gorgon Halo', announced May 2026) lifts the ceiling to 192 GB with OEM systems from Q3 2026.
Siblings on the same silicon
HP
Z2 Mini G1a
The corporate Strix Halo — a 2.7-litre workstation with tier-1 warranty, in stock in the UK at £2,663.99.
- Memory
- 128 GB
- Bandwidth
- 256 GB/s
- AI perf
- 40-CU RDNA 3.5
£2,663.99 inc VAT
hp.com/gb-en list (128 GB/2 TB) — seen at £2,397 with promo codes
Framework
Framework Desktop
The community favourite — the best-documented local-LLM box on the platform, from the repairability company.
- Memory
- 128 GB
- Bandwidth
- 256 GB/s
- AI perf
- 40-CU RDNA 3.5
~$2,851 (128 GB + 1 TB)
US verified Mar 2026 · UK GBP via configurator (~£2.2–2.5k expected) · rose ~$460 in the RAM crisis
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 EVO-X2 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.