Verdict
All guidesMethodologyAboutContact
Head-to-head · Best AI Workstations

HP Z6 G5 A vs NVIDIA DGX Spark

Which is the better buy? Side-by-side on rating, price, strengths, and watch-outs — with the published ratings we averaged to get there.

The short answer

NVIDIA DGX Spark comes out ahead by a narrow margin (4.5 vs 4.6). The gap is mostly about Best for local-LLM developers — CUDA-native 128 GB dev box — read the strengths below before deciding.

HP Z6 G5 A
Ranked #3 in Best AI Workstations
HP Z6 G5 A
$5,499as of Apr 25

The HP Z6 G5 A is the smallest Threadripper Pro OEM workstation on the market and the rational mid-tier pick under HP's flagship Z8 Fury G5. Reviewers across PCMag, AnandTech, StorageReview, Phoronix, and DEVELOP3D consistently praised its build quality, toolless serviceability, and 96-core CPU ceiling — StorageReview gave it their 'highest recommendation for a high-end tower workstation.' For local-LLM use, configurations with 1–3 RTX 6000 Ada GPUs (48 GB VRAM each at ~960 GB/s) deliver in the 25–40 tokens/sec range on Llama-3-70B Q4 single-GPU and substantially more with multi-GPU tensor parallelism. Note that none of the published professional reviews ran formal Llama-3 70B Q4 benchmarks, so LLM-specific performance numbers here are from single-GPU norms rather than published HP Z6 measurements specifically.

Strengths
  • — Smallest Threadripper Pro OEM tower on the market — compact 4U chassis with built-in handle
  • — AMD Ryzen Threadripper Pro 7000 WX-Series scales from 12 to 96 cores at the same chassis price floor
  • — Toolless serviceability, modular interior, ECC DDR5 — enterprise pedigree at mid-tier pricing
Watch-outs
  • — 95°C all-core CPU thermals reported under sustained load (StorageReview)
  • — Pricing scales steeply — 96-core configs push $18,000+
  • — No published Llama 70B Q4 tokens/sec figures in mainstream reviews — LLM-specific benchmarking is thin
$5,499 · Buy at hp.comFull review →
NVIDIA DGX Spark
Higher ratedRanked #2 in Best AI Workstations
NVIDIA DGX Spark
$4,699as of Apr 25

The NVIDIA DGX Spark is the productized version of Project DIGITS — a 150 mm cube housing the GB10 Grace Blackwell Superchip, 128 GB unified LPDDR5x, and the full CUDA AI stack out of the box. Tom's Hardware called it 'a well-rounded toolkit for local AI'; ServeTheHome called it 'must-have for AI developers'; LMSYS published the most thorough independent benchmarks. The 128 GB unified-memory ceiling is the headline feature: it loads models that would otherwise need a $30K+ multi-GPU rig. The catch is bandwidth-limited decode — LMSYS measured Llama-3.1 70B FP8 at 2.7 tokens/sec single-batch, while GPT-OSS 120B (MoE, ~17B active) hits ~14.5 tokens/sec per ServeTheHome. Best understood as a CUDA-native development box for buyers who need to iterate on big-model code without renting cloud GPUs.

Strengths
  • — 128 GB unified LPDDR5x memory — fits 70B FP8 / 120B Q4 / 405B with two clustered units
  • — Full CUDA + NVIDIA AI stack preinstalled; the most polished local-AI dev box on the market
  • — Compact 150 mm cube, 240 W max — fits any desk, runs cool and quiet
Watch-outs
  • — 273 GB/s LPDDR5x bandwidth caps decode tok/s on dense large models — 70B FP8 measures ~2.7 tok/s on a single unit
  • — Linux-only, no Windows or gaming use; specialist hardware for AI developers
  • — Price raised from $3,999 to $4,699 in February 2026 due to memory supply
$4,699 · Buy at nvidia.comFull review →

How they stack up

HP Z6 G5 A

The HP Z6 G5 A is the mid-tier sweet spot in this lineup. Versus the HP Z8 Fury G5 (its flagship sibling), it's a smaller chassis with the same Threadripper Pro CPU family at a noticeably lower entry price — trading the Z8's 4-GPU ceiling for a 3-GPU ceiling and a more desk-friendly footprint. Versus the Puget Genesis II, it offers similar build pedigree without Puget's bespoke configurator and handpicked components, at a meaningfully lower starting price. Versus the DGX Spark, it's a different class of machine — the HP Z6 G5 A is a multi-GPU general workstation, the Spark is a single-purpose 128 GB unified-memory dev box. Pick the HP Z6 G5 A when you need both AI horsepower and traditional workstation workloads (rendering, simulation, multi-app productivity) on the same machine.

NVIDIA DGX Spark

The DGX Spark is the cheapest path to 128 GB of CUDA-addressable unified memory anywhere on the market. Versus the GMKtec EVO-X2 ($1,699) or Beelink GTR9 Pro ($2,000), it's roughly 2.5x the price but offers the full NVIDIA software stack the Strix Halo boxes can only approximate via ROCm or Vulkan. Versus the Puget Genesis II ($10K+), it's a single-purpose dev box — no multi-display creative workflow, no gaming, no general workstation duty. Pair two Sparks via the ConnectX-7 networking and you get 405B-class model coverage at roughly $9,400, the cheapest legal path to that ceiling.

Specs side-by-side

SpecHP Z6 G5 ANVIDIA DGX Spark
CPUAMD Ryzen Threadripper Pro 7000 WX-Series (12–96 cores)20-core Arm (10x Cortex-X925 + 10x Cortex-A725)
GPUUp to 3x dual-height pro GPUs (RTX A6000, RTX 6000 Ada)NVIDIA GB10 Grace Blackwell Superchip
RAMUp to 1 TB DDR5-5600 ECC (8 channels)128 GB unified LPDDR5x
StorageHP Z Turbo NVMe (multiple M.2 + bays)4 TB NVMe M.2 (user-replaceable, self-encrypting)
Memory Bandwidth~358 GB/s system; ~960 GB/s per RTX 6000 Ada VRAM273 GB/s
Form FactorCompact 4U tower (169 x 465 x 445 mm, built-in handle)Compact desktop (150 mm cube, 1.2 kg)
← See the full ranking of best ai workstations