HP ZGX Fury Specs: 748 GB GB300 Tower, No US Price, 200-240 V
The HP ZGX Fury has been orderable since Sept. 9 with no US price; HP Japan lists it from 28,490,000 yen. The only lab data is a two-tower cluster run, and StorageReview's solo review is under way.
Local AI DeskEdited by Ege Erten
5 min readAI-assisted
Follows our story: NVIDIA DGX Station Review: 748 GB, 1.6 kW and a $160,195 Dell List Price

The HP ZGX Fury G1n AI Station, as shown in HP Japan's press photo library.
Image: HP
Source · Press image from the HP Japan newsroom, used for news reporting. · License
The HP ZGX Fury AI Station, a GB300 tower with 748 GB of memory, needs a dedicated 200–240 V AC circuit and will not run from a standard 120 V household outlet, HP's US product page says.
HP opened orders on Sept. 9 and its US page still shows no price. The only list price is in Japan: HP Japan announced the ZGX Fury G1n on Oct. 26 and began sales on Oct. 27, from 28,490,000 yen including tax. Our Nov. 1 DGX Station review already covered that price and the platform's two-tower cluster test. This piece sets out what HP's own pages add, what the one lab run shows about the HP unit, and what is still missing. We found no published measurement of a single ZGX Fury, so this is not a review, and Voltgrid did not test the machine.
HP ZGX Fury Specs: What HP's Page Lists

HP states the ZGX Fury's 748 GB as 252 GB of HBM3e plus 496 GB of LPDDR5x; the ZGX Nano has 128 GB (HP figures).
Illustration (AI-generated)
The silicon is NVIDIA's DGX Station design. HP lists a 72-core NVIDIA Grace Neoverse V2 CPU, soldered to the board, with four 128 GB SOCAMM modules of LPDDR5x: 496 GB at 396 GB/s. The Blackwell Ultra GPU carries 252 GB of HBM3e at 7.1 TB/s. HP adds the two pools and calls the result 748 GB of unified memory, and it rates the GB300 at up to 20 petaFLOPS FP4, a figure HP does not label as sparse or dense; NVIDIA's DGX Station table gives 20 with sparsity and 15 without.
Storage is split in two. Two M.2 drives on PCIe Gen5 hang off the CPU and hold the operating system as a software RAID 1 mirror. Two more on PCIe Gen6 come off the ConnectX-8 network card and form a software RAID 0 data volume, HP says. StorageReview reported on Sept. 9 that the data drives are 2 TB or 4 TB self-encrypting NVMe, chosen at purchase. That is the Gen5-plus-Gen6 layout ASUS lists for its ET900N G3, not the four Gen5 slots on NVIDIA's own listing.
The box is built for a rack as much as a desk. It runs as a tower or in a standard 5U rack slot on included rails, HP says, with liquid cooling and front-to-back airflow so that racking it does not block ventilation. The ConnectX-8 brings two QSFP112 ports at 400 Gbps each for linking a second unit, plus a 10 Gbps RJ-45 port and a separate management controller with its own 1 Gbps port. Power enters through a C20 inlet. The GB300 drives no monitor itself; HP offers an optional NVIDIA RTX PRO card for displays, StorageReview reported.
Software is Linux first. The ZGX Fury ships with Ubuntu 24.04 LTS, NVIDIA's AI developer tools and two HP layers: Z Runtime, a preinstalled command-line tool that pulls and serves models, and Z Toolkit, a free VS Code extension. In its Sept. 9 release, HP said the machine is certified for Red Hat Enterprise Linux. A Red Hat AI Factory with NVIDIA platform on the Fury is planned, with timing and supported configurations still to be announced.
What the 20 PFLOPS FP4 Rating Means
HP rates the ZGX Fury at up to 20 petaFLOPS of FP4 and puts no sparsity label on that figure. Structured sparsity lets hardware skip half the weights in a matrix, so vendors that count it quote twice the dense rate. NVIDIA's table for its DGX Station lists 20 PFLOPS of FP4 with sparsity and 15 without.
A peak rating is not a speed. Local inference is usually limited by how fast weights stream from memory, and the Fury has two tiers: 252 GB of HBM3e at 7.1 TB/s and 496 GB of LPDDR5x at 396 GB/s, HP says. StorageReview's 5,018 tokens per second came from two towers running one model, not from the 20 petaFLOPS rating.
What Models the ZGX Fury Can Hold
HP pitches the ZGX Fury as a shared machine for a team, not a personal PC. It says the Fury hosts models of up to 1 trillion parameters and fine-tunes models up to the 100-billion-parameter class, with a footnote that its model harness quantizes at FP4. The ZGX Nano has 128 GB at 273 GB/s; HP's own FAQ says teams prototype on the Nano and move the same workflow to the Fury for larger models, more throughput and more users.
The memory split sets the speed. A model that fits in the 252 GB of HBM3e reads its weights at 7.1 TB/s. Anything larger spills into the LPDDR5x at 396 GB/s, about an eighteenth of that rate by Voltgrid's arithmetic on HP's two figures; HP states no ratio. The 748 GB decides which models load at all; the 252 GB decides which stay fast.
What the Only Lab Data Shows
The one independent measurement involving the HP is a cluster run. StorageReview, on Oct. 5, linked an ASUS ET900N G3 and a ZGX Fury with two 400G direct-attach copper cables. GPUDirect RDMA reached 392 Gb/s on each rail, and NCCL collectives hit 98 GB/s on broadcast, reduce and send/receive. With GLM-5.3 in NVFP4 split across both GPUs under vLLM 0.27.1, the pair reached 5,018 output tokens per second at 256 streams, against 188 at 32 streams on one tower. The two points are not a like-for-like multiple, because the stream counts differ, and the split used tensor parallelism with an MTP draft setting that StorageReview chose.
The HP had its own job in one test. For GLM-5.3 Flash, the ASUS processed the prompts and the HP generated the tokens. With 512-token prompts and answers on vLLM 0.28.1 RC1, output at 32 streams rose from 828 tokens per second on one Station to 2,721 on the pair. Each point was a single pass. The single-tower baseline in every chart was the ASUS, so none of these figures is a solo HP result, and they say nothing about HP's cooling, clocks or noise.
What Is Still Unknown
There is no US, UK or EU list price. HP's Oct. 7 release told buyers of the Windows version to contact a local sales representative. For scale, Dell.com listed its GB300 tower at $160,195 in its default configuration with an RTX PRO 2000, and Central Computers listed the ASUS from $111,999.99 as a quote-only build with two 2 TB drives, but the builds differ and neither figure says anything about HP's price.
HP's page gives no power supply rating, weight, dimensions or noise level; it refers buyers to the QuickSpecs for full power specifications. NVIDIA rates the platform at 1,600 W of total system power on a 20 A circuit, and no lab has measured an HP unit at the wall.
Windows is planned, not shipping. HP first showed Windows running on the ZGX Fury at a Microsoft event on Oct. 7, and HP Japan says Windows support is planned from December 2026. StorageReview notes that whether serving stacks will run natively on Windows or under WSL is still open.
StorageReview's desktop ranking page says the ZGX Fury is in its lab and the solo review is under way, and that it will rank the machine once its own bench numbers are in. The page gives no date for that statement. That review, a US price and the December Windows release are the next milestones.
How it compares
Full table| Product | Memory | Bandwidth | AI compute | Measured speed | Price |
|---|---|---|---|---|---|
| ASUS ExpertCenter Pro ET900N G3ASUS (NVIDIA DGX Station GB300 platform) | 748 GB coherent: 252 GB HBM3e (GPU) + 496 GB LPDDR5X (CPU)[1] | 7.1 TB/s (GPU HBM3e); 396 GB/s (CPU LPDDR5X)[1] | up to 20 PFLOPS AI performance (ASUS gives no precision or sparsity label) (sparse)[2] | GPT-OSS-120B MXFP4, vLLM, 512/512, 1-128 streams: 9,280 aggregate output tok/s at 128 streams; 5,023 on 8,192/1,024 (StorageReview, ASUS-supplied unit, no add-in GPU)[3] | $111,999.99 at Central Computers, US (model ET900N G3-2SSD-V2, request a quote, unavailable online, read Nov 4, 2026); no ASUS list price[4] |
| HP ZGX Fury AI StationHP | 252 GB HBM3e GPU + 496 GB LPDDR5x CPU (4x 128 GB SOCAMM); 748 GB unified[5] | 7.1 TB/s (GPU HBM3e); 396 GB/s (CPU LPDDR5x)[5] | up to 20 petaFLOPS FP4 (HP gives no sparsity label) (sparse)[5] | Two-unit cluster with an ASUS ET900N G3 over two 400G DACs: GLM-5.3 NVFP4 5,018 output tok/s at 256 streams, 512/512, vLLM 0.27.1 TP2 with MTP (StorageReview); no single-unit HP result[6] | Japan list price from 28,490,000 yen incl. tax (sales from Oct 27, 2026); no US list price published[7] |
| NVIDIA DGX StationNVIDIA (systems built by ASUS, Dell, GIGABYTE, HP, MSI, Supermicro and Exxact) | 748 GB coherent: 252 GB HBM3e (GPU) + 496 GB LPDDR5X (CPU)[8] | GPU HBM3e 7.1 TB/s; CPU LPDDR5X 396 GB/s; NVLink-C2C 900 GB/s between them[8] | 15 PFLOPS FP4 (without sparsity)[8] | gpt-oss-120b MXFP4, stock vLLM, 512/512 chat workload: 9,280 output tok/s at 128 streams (5,023 on 8,192/1,024 prompts), single GB300 (StorageReview, ASUS ExpertCenter Pro ET900N G3)[9] | No NVIDIA list price; lowest US partner listing read Nov 4, 2026: Exxact Valence VWS-158270643 from $98,450 (Tom's Hardware found $94,930 in Aug); Dell Pro Precision with GB300 lists $160,195[10] |
| NVIDIA RTX PRO 6000 Blackwell Workstation EditionNVIDIA (sold through PNY, TD SYNNEX and workstation makers) | 96 GB[11] | 1,792 GB/s[11] | 4,000 AI TOPS (FP4, with sparsity) (sparse)[11] | gpt-oss-120b MXFP4, stock vLLM, 512/512 chat workload: 3,064 output tok/s at 128 streams (1,169 on 8,192/1,024 prompts); Server Edition at 600W (StorageReview)[9] | $16,000 on NVIDIA's US marketplace (Workstation Edition), listed out of stock[12] |
| Apple Mac Studio (M5 Ultra)Apple | up to 512 GB unified (96 GB on 30-core M5 Ultra; 256 GB or 512 GB on 36-core/80-core M5 Ultra)[13] | 1.2 TB/s[13] | — | Qwen 3.8 27B 4-bit MLX in LM Studio: about 55 tok/s on M5 Ultra (36-core, 256 GB) vs about 40 on M3 Ultra (BGR)[14] | $22,299 minimum with 512 GB ($6,799 36-core CPU/80-core GPU M5 Ultra config plus $15,500 memory upgrade; Oct 28, 2026)[15] |
| NVIDIA RTX Spark (N1X) PCsNVIDIA (MediaTek collaborated on the custom CPU design) | Up to 128 GB LPDDR5X (20-core CPU / 6,144-core GPU version); up to 64 GB (18-core / 5,120-core laptop version)[16] | Up to 300 GB/s: LPDDR5X-9400 on a 256-bit bus (NVIDIA's product page lists no bandwidth)[17] | — | gpt-oss:20b via Ollama in a Codex CLI coding agent, 64K context: 58.5 tok/s generation, 3 of 4 tasks passed (Dell XPS 16, 20-core N1X, 64 GB; StorageReview)[18] | Laptops from $2,599; pricing and configurations vary by region[19] |
| AMD Ryzen AI Halo (Ryzen AI Max+ PRO 495, 192 GB)AMD | 192 GB LPDDR5X (256-bit)[20] | 273 GB/s[20] | — | — | $7,299.99 for the Linux version (released the week of Oct 12, 2026); Windows version coming soon[20] |
| ASUS Ascent GX10ASUS | 64 GB or 128 GB LPDDR5X-8533 unified[21] | 273 GB/s (8533 MT/s), as listed by StorageReview; ASUS's spec page states no bandwidth[22] | Up to 1 PFLOP FP4 (GB10; NVIDIA's footnote: theoretical, using the sparsity feature). ASUS's pages say 1 petaFLOP / FP4 with no sparsity label (sparse)[23] | gpt-oss-120b, vLLM online serving, equal ISL/OSL: 70 tok/s at batch 1 rising to 680 tok/s aggregate at batch 64 (StorageReview)[22] | $5,999 (128 GB, 1 TB, GX10-GG0010BN, ASUS eShop US, Oct 30, 2026)[24] |
| Dell Pro Precision with GB10 (model FCM1253; launched as Dell Pro Max with GB10)Dell | 128 GB LPDDR5x up to 8533 MT/s onboard (64 GB option also listed)[25] | 273 GB/s (128 GB LPDDR5x, 256-bit)[26] | — | gpt-oss-120b, vLLM online serving, equal ISL/OSL: 68.57 tok/s at batch 1 rising to 670.83 tok/s aggregate at batch 64 (StorageReview)[27] | $7,967 (128 GB, 2 TB Gen4 QLC SSD default config, Dell.com US, read Nov 2, 2026 16:40 UTC)[25] |
| Framework Desktop (Ryzen AI Max+ PRO 495, 192 GB)Framework | 192 GB LPDDR5X-8533 soldered[28] | 273 GB/s (256-bit)[28] | — | — | $6,799 (Max+ PRO 495 192 GB DIY Edition pre-order, Nov 3, 2026; storage not included)[29] |
| GIGABYTE AI TOP ATOMGIGABYTE | 128 GB unified system memory[30] | 273 GB/s (LPDDR5x-8533, 256-bit)[31] | — | gpt-oss-120b, vLLM online serving, equal ISL/OSL: 60.71 tok/s at batch 1 rising to 649.62 tok/s aggregate at batch 64 (StorageReview)[31] | $7,312.00 advertised at CDW US against a $7,999.99 list (128 GB, 4 TB, ATAGB10-9000; read Nov 2, 2026)[32] |
| GMKtec EVO-X5 Pro AI Mini PCGMKtec | 192 GB LPDDR5X-8533 onboard (8 x 24 GB, non-upgradeable); up to 160 GB allocatable as VRAM[33] | Up to 273 GB/s[34] | — | GMKtec Laboratory table: GLM-5.3-Flash UD-IQ4_XS (146.05 GB) decode 17.79 tok/s, prefill 70.18, TTFT 14.71 s; Qwen3.8-Flash-Next ROCmFP4-FAST (93.48 GB) decode 45.08 tok/s[33] | $6,799 (192 GB/2 TB) and $7,099 (192 GB/4 TB), GMKtec US store, Nov 2, 2026[33] |
| HP ZGX Nano AI Station (G1n)HP | 128 GB coherent unified system memory[35] | 273 GB/s (128 GB LPDDR5x)[36] | GB10: up to 1 PFLOP FP4, theoretical, using the sparsity feature; HP and reviewers list 1,000 TOPS FP4 (sparse)[23] | gpt-oss-120b, vLLM online serving, equal ISL/OSL: 69.6 tok/s at batch 1 rising to 722.9 tok/s aggregate at batch 64 (StorageReview)[37] | $7,399.00 at the HP Store for the 4 TB model; $4,759 on Amazon (Notebookcheck, review updated Oct 5, 2026)[38] |
| Lenovo ThinkStation PGXLenovo | 128 GB LPDDR5X unified (256-bit)[39] | 273 GB/s[39] | — | — | about $4,100 for the 1 TB configuration (Mar 2026)[40] |
| MSI EdgeXpert (MS-C931)MSI | 64 GB or 128 GB LPDDR5x unified (256-bit)[41] | 273 GB/s[41] | 1 PFLOPS FP4 (sparse)[41] | — | No list price; MSI sells via quote (Get a quote)[41] |
| NVIDIA DGX SparkNVIDIA (partner GB10 systems from Acer, ASUS, Dell, GIGABYTE, HP, Lenovo and MSI) | 128 GB; a 64 GB configuration is sold only through OEM partners[23] | 273 GB/s[23] | Up to 1 PFLOP FP4 (theoretical, using the sparsity feature) (sparse)[23] | Qwen 3.8 27B (dense) with multi-token prediction: decode of about 20 tokens/s, with time-to-first-token staying interactive out to the full 262K context.[42] | $6,950.00 (128 GB, 4 TB) on NVIDIA's enterprise marketplace with an Add To Cart button in a copy cached 07:59 UTC on 2026-10-31; stock unverified (listed out of stock on 2026-10-21); 64 GB partner systems from $4,999[43] |
| NVIDIA DGX Spark 64 GBNVIDIA (sold by Acer, ASUS, Dell, Gigabyte, HP and MSI) | 64 GB LPDDR5X unified (two units pool 128 GB over ConnectX-7)[44] | 273 GB/s (256-bit; NVIDIA spec table covers both 64 GB and 128 GB)[23] | 1 PFLOPS FP4 (sparse; NVIDIA spec table covers both memory configurations)[23] | — | from $4,999 (partner systems, Oct 2026; no Founders Edition)[44] |
| Sipeed MaixStationSipeed | 128 GB LPDDR5X (256-bit)[20] | 273 GB/s (Sipeed claim)[20] | 256 TOPS FP4 (native FP4, no sparsity; Axera/Sipeed claim)[20] | — | about $3,999 expected for 128 GB, with Kickstarter launch discounts (Sipeed to Tom's Hardware, Oct 2026)[20] |
| AMD Ryzen AI Halo Developer PlatformAMD | 128 GB LPDDR5x-8000 unified[45] | 256 GB/s[45] | — | gpt-oss-120b, vLLM online serving, equal ISL/OSL 256/256: 222 tok/s aggregate at batch 64 vs 701 on DGX Spark (StorageReview)[46] | $3,999 with 2 TB SSD (Jul 2026)[46] |
| Framework Desktop (Ryzen AI Max+ 395)Framework | 128 GB LPDDR5x-8000 soldered (32 GB and 64 GB options)[47] | 256 GB/s[48] | — | gpt-oss-120b MXFP4 in LM Studio (out-of-the-box setup): roughly 45 tok/s (ServeTheHome; Ryzen AI Max+ 395, 128 GB, Windows 11 Pro)[49] | $3,449 (DIY Edition, Max+ 395 with 128 GB, system only; out of stock, Nov 3, 2026)[50] |
| Acer Veriton GN100 AI Mini WorkstationAcer | 128 GB LPDDR5x coherent unified[51] | — | — | gpt-oss-120b, vLLM, total token throughput: prefill heavy 303.61 to 2,777.56 tok/s; decode heavy 38.38 to 292.41 tok/s (batch 1 to 64, StorageReview)[52] | from $3,999 in North America (announced Sep 3, 2025; prices vary by region)[51] |
| ASRock Industrial AI BOX-A395ASRock Industrial | 128 GB LPDDR5X-8000 soldered[53] | — | — | AgentSTH V7, 32 threads: 15% ahead of the next-closest Ryzen AI Max+ 395 box (single-core within 1%); estimated SPEC CPU2026 int rate: top of a 44% spread across four 395 systems; measured by ServeTheHome, Ubuntu 26.04[54] | — |
| Dell Pro Precision with GB300 (model FCT6263; announced as Dell Pro Max with GB300)Dell | 252 GB HBM3e GPU + 496 GB LPDDR5X CPU[55] | — | — | — | $160,195.00 (Dell.com US default config FCT6263 with RTX PRO 2000 display GPU, read Nov 4, 2026); RTX PRO 6000 Workstation Edition option adds $13,560[55] |
| GMKtec EVO-X2 AI Mini PCGMKtec | 64 GB or 128 GB onboard LPDDR5X-8000 (non-upgradeable)[56] | — | — | Llama 3.3 70B Q6_K Instruct: 3.7-3.8 tok/s generation fully on the Radeon 8060S (about 2 tok/s CPU-only); Qwen3-32B 10.4-10.5 tok/s; 128 GB unit, short prompt, tool not named (ServeTheHome)[57] | $3,499.99 (128 GB/1 TB) and $3,649.99 (128 GB/2 TB), each against $3,999.99; $2,199.99 for 64 GB/1 TB (GMKtec US store, read Nov 4, 2026)[56] |
| HP Z2 Mini G1a WorkstationHP | up to 128 GB unified, up to 96 GB assignable to the GPU, up to 8000 MT/s[58] | — | up to 50 NPU TOPS[58] | gpt-oss-120b GGUF (63.39 GB, full GPU offload) in LM Studio: close to 40 tok/s (StorageReview)[59] | $4,781 list for StorageReview's review configuration; about $3,300 street (B&H top spec with 2 TB SSD at $3,342.65, Aug 2025)[59] |
| Microsoft Surface RTX Spark Dev BoxMicrosoft | 128 GB unified memory[60] | — | up to 1 PFLOP of AI compute (Microsoft; precision and sparsity not stated on the page) (sparse)[60] | — | From $5,999.99 (Microsoft Store US, pre-order, read Nov 1, 2026)[60] |
| MSI XpertStation WS300MSI (NVIDIA DGX Station platform) | 252 GB HBM3e GPU + 496 GB LPDDR5X CPU (748 GB coherent)[61] | — | — | StorageReview vLLM: gpt-oss-120b MXFP4 9,294 output tok/s at 128 streams (512/512), 5,038 on 8,192/1,024; MiniMax M3 NVFP4 1,041 at 32 streams with EAGLE3-GQA at a forced acceptance length of 3 (best-case)[61] | $99,999.00 'Start From' at Newegg US (model WS300T60L, 2x 2 TB NVMe RAID1), sold by Newegg as a request-a-quote listing (read Nov 4, 2026)[62] |
| Tenstorrent TT-QuietBox 2 (Blackhole)Tenstorrent | 128 GB GDDR6 on the accelerators + 256 GB DDR5-5600 system memory[63] | — | — | — | $9,999 (Oct 30, 2026)[63] |
| WiCi One (RTX 5060 Ti edition)WiCi | 16 GB GDDR7 VRAM; 1 TB NVMe 4 SSD for TurboStream weight paging (WiCi's claim: NVMe used as working VRAM). System RAM not stated[64] | — | — | — | $2,599 list (WiCi shop, Oct 2026); a $600-off code (WICIWELCOME) is applied at checkout. Early-access price was $1,999 (Notebookcheck, Engadget, Sep 2026)[64] |
| WiCi One (RTX 5090 edition)WiCi | 32 GB GDDR7 VRAM; 4 TB NVMe 5 SSD for TurboStream weight paging (WiCi's claim: NVMe used as working VRAM). System RAM not stated[64] | — | — | — | $9,999 list (WiCi shop, Oct 2026); a $600-off code (WICIWELCOME) is applied at checkout[64] |
Italic values are reported by an outlet or measured by an independent lab; the rest come from the maker. Measured speeds use different models, quantizations and tools, so read each with its source rather than against the others. Voltgrid News does not test hardware itself.
Sources (64)
- [1] ASUS
- [2] ASUS
- [3] StorageReview
- [4] Central Computers
- [5] HP
- [6] StorageReview
- [7] HP Japan
- [8] NVIDIA
- [9] StorageReview
- [10] Exxact
- [11] NVIDIA
- [12] NVIDIA
- [13] Apple
- [14] BGR
- [15] MacRumors
- [16] NVIDIA
- [17] ServeTheHome
- [18] StorageReview
- [19] NVIDIA
- [20] Tom's Hardware
- [21] ASUS
- [22] StorageReview
- [23] NVIDIA
- [24] ASUS eShop
- [25] Dell
- [26] HotHardware
- [27] StorageReview
- [28] Framework Blog
- [29] Framework
- [30] GIGABYTE (PR Newswire)
- [31] StorageReview
- [32] CDW
- [33] GMKtec
- [34] GMKtec
- [35] HP
- [36] HotHardware
- [37] StorageReview
- [38] Notebookcheck
- [39] Lenovo PSREF
- [40] ServeTheHome
- [41] MSI
- [42] Tom's Hardware
- [43] NVIDIA Marketplace
- [44] NVIDIA Blog
- [45] AMD
- [46] StorageReview
- [47] Framework
- [48] Framework Blog
- [49] ServeTheHome
- [50] Framework
- [51] Acer
- [52] StorageReview
- [53] ServeTheHome
- [54] ServeTheHome
- [55] Dell
- [56] GMKtec
- [57] ServeTheHome
- [58] HP
- [59] StorageReview
- [60] Microsoft
- [61] StorageReview
- [62] Newegg
- [63] Tenstorrent
- [64] WiCi
Sources
- 01HP ZGX Fury: AI-Powered WorkstationHP · hp.com
- 02HP Extends Data Center AI Architecture to the EdgeHP · hp.com
- 03HP ZGX Fury Is Now Orderable: GB300 Superchip, 748GB Unified Memory, and a Red Hat AI Factory Plan for the EdgeStorageReview · storagereview.com
- 04日本HP、高度なAI開発に特化した高性能ワークステーション「HP ZGX Fury G1n AI Station」を発表HP Japan · jp.ext.hp.com







