ASUS Ascent GX10 2025, NVIDIA GB10 Grace Blackwell Superchip, 128GB, 2TB SSD, Stellar Gray

★★★★★ 5.0 (2)

NVIDIA GB10 Grace Blackwell Superchip과 128GB LPDDR5x 통합 메모리를 탑재하여 소형 메탈 섀시 안에서 페타플롭스급 AI 연산 성능을 제공합니다. NVIDIA DGX Spark 플랫폼 기반의 이 제품은 OpenClaw 및 NemoClaw 같은 프레임워크를 지원하며 온디바이스 추론과 샌드박스 실행 환경을 갖춘 것이 특징입니다. 보안이 중요한 장기 실행형 에이전트 워크플로우를 구축하는 AI 개발자에게 적합합니다.

CPU 20-Core: Up to 3.25 GHz Cortex A725 (10 Cores)Up to 3.8 GHz Cortex X925 (10 Cores)
RAM 128 GB
Storage 2 TB
GPU NVIDIA GB10 Grace Blackwell Superchip
Form Factor mini
OS Ubuntu
ASUS Ascent GX10 2025, NVIDIA GB10 Grace Blackwell Superchip, 128GB, 2TB SSD, Stellar Gray desktop
74 종합 점수
다른 국가에서도 구매 가능:

요약

The 30-Second Version

The ASUS Ascent GX10 is a mini AI workstation with best-in-class 128GB unified RAM and a Grace Blackwell Superchip. It's excellent for local AI development but useless for gaming. Worth it for AI devs who can find it at the lower end of the price range.

Pros & Cons

장점

  • 128GB of unified RAM is best-in-class for local AI workloads 99th
  • Tiny footprint with low noise and power draw 90th
  • 2TB NVMe SSD and 10GbE networking are serious specs 84th
  • Runs Ubuntu with OpenClaw and NemoClaw support out of the box 75th

단점

  • Gaming performance is basically nonexistent at 22/100
  • Only 4 USB-C ports and one HDMI, port selection is thin
  • Reliability scores are below average for the price
  • No Windows option, you're locked into Ubuntu

사용자 의견

The Word on the Street

5.0/5 (2 reviews)
👍 Owners consistently praise the small footprint, low noise, and low power consumption while still handling large models and contexts.
👍 Buyers say it's exactly what they needed for local AI work without cloud dependency.

근거 자료

Performance

The GB10 Grace Blackwell Superchip is the star here. It pairs a 20-core CPU with an integrated GPU and 128GB of unified memory, which means large models and long contexts fit without breaking a sweat. Our database puts the RAM at the 99th percentile, storage at the 90th, and GPU at the 84th. That's best-in-class memory and a standout graphics setup for this category. The CPU lands at 72nd percentile, which is solid but not the main event. Port selection is mediocre at 39th percentile, and reliability sits at 37th, which is a weak spot for a machine this expensive. The 2TB NVMe SSD is fast and roomy, and the ConnectX-7 SmartNIC with 10GbE plus Wi-Fi 7 means network throughput won't be your bottleneck.

Performance Percentiles

CPU 52.5
GPU 84.2
RAM 98.9
포트 38.6
저장 공간 90.2
신뢰성 37.4
사용자 평판 75.1

Specifications

Full Specifications

Processor

CPU 20-Core: Up to 3.25 GHz Cortex A725 (10 Cores)Up to 3.8 GHz Cortex X925 (10 Cores)
Cores 20
Frequency 3.3 GHz

Graphics

GPU NVIDIA GB10 Grace Blackwell Superchip
Type Integrated

Memory & Storage

RAM 128 GB
RAM Generation DDR5
Storage 2 TB
Storage Type NVMe SSD

Build

Form Factor mini
Weight 3.2 kg / 7.0 lbs

Connectivity

USB-C Ports 4
HDMI HDMI 2.1
Wi-Fi Wi-Fi 7
Bluetooth Bluetooth 5.4
Ethernet 10GbE LAN

System

OS Ubuntu

vs Competition

Against the Apple Mac Studio M5 Max, the GX10 is more specialized. The Mac Studio is a better all-rounder with stronger general computing and creative app support, but it can't touch 128GB of unified memory at this price. The Lenovo Legion Tower 5a Gen 10 and Alienware Aurora ACT1250 are gaming desktops first, so they'll crush the GX10 in games but fall flat for large model inference. The Dell ECT1250 and HP OMEN GT16-0364 are similar stories, more general purpose but less focused. If you need a desktop that does everything, buy one of those. If you need a local AI box, the GX10 is in a different lane.

Spec ASUS Ascent GX10 Apple Mac Studio M5 Max Lenovo T Series Towers Legion Tower 5a Gen 10 HP OMEN GT16-0364 Dell Area-51 CLX SET TGMSETRTU5204BM
CPU 20-Core: Up to 3.25 GHz Cortex A725 (10 Cores)Up to 3.8 GHz Cortex X925 (10 Cores) Apple M5 Max AMD Ryzen 7 7800X3D AMD Ryzen 7 8700F Intel Core Ultra 9 285K Intel Core i9 14900KF
RAM (GB) 128 64 32 32 32 64
Storage (GB) 2048 4096 4096 1024 2048 8192
GPU NVIDIA GB10 Grace Blackwell Superchip Apple (40-Core) NVIDIA GeForce RTX 5070 Ti NVIDIA GeForce RTX 5070 NVIDIA GeForce RTX 5080 NVIDIA GeForce RTX 5070
Form Factor mini mini mid-tower mid-tower desktop mid-tower
Psu W - 480 850 850 1500 850
OS Ubuntu macOS Windows 11 Home Windows 11 Home Windows 11 Home Windows 11 Home
Compare Compare Compare Compare Compare
제품 CPUGPURAM포트저장 공간신뢰성사용자 평판
ASUS Ascent GX10 52.584.298.938.690.237.475.1
Apple Mac Studio M5 Max Compare 95.883.698.489.198.299.381.4
Lenovo T Series Towers Legion Tower 5a Gen 10 Compare 73.593.178.188.498.270100
HP OMEN GT16-0364 Compare 69.889.287.39853.77099.1
Dell Area-51 Compare 98.295.990.799.590.27058.4
CLX SET TGMSETRTU5204BM Compare 9589.296.584.299.211.694.8

가격

Value & Pricing

Pricing is all over the place. In the US you'll see USD 6,999 to USD 8,700 across two listings, and in Canada it ranges from CAD 6,579 to CAD 9,499. That's a huge spread, so shop around before you buy. The best US deal is at the lower end of that range, and the best Canadian price is CAD 6,579. For a machine with this much unified memory and a Grace Blackwell chip, the lower prices feel reasonable for AI developers. At the top end, you're paying a premium for availability, not extra hardware.

최저 US$8,700 소매점 1곳, 가격 1개
Amazon 가격 1개 최저 US$8,700

Price History

US$4,000 US$5,000 US$6,000 US$7,000 US$8,000 US$9,000 US$10,000 6월 18일8월 14일10월 4일11월 6일 US$6,999

더 보기

Overview

The ASUS Ascent GX10 is a tiny AI workstation that packs the NVIDIA GB10 Grace Blackwell Superchip into a stackable metal box. It's built for developers running local inference and agentic workflows, not for gaming or general desktop duty. With 128GB of unified LPDDR5x RAM and a 2TB NVMe SSD, it's aimed squarely at people who want petaflop-scale AI performance without renting cloud GPUs.

This thing is small, quiet, and surprisingly power-efficient for what it does. But it's also expensive, runs Ubuntu, and won't help you if you need a general-purpose PC. If you're an AI developer who wants a private, on-device sandbox for large models, this is a very specific tool that does its job well.

Common Questions

Q: Can I use this as a regular desktop PC?

You can, but it's not great at it. It runs Ubuntu, has limited ports, and gaming performance is very weak. It's really built for AI development and inference.

Q: Does it support Windows?

No, it ships with Ubuntu and NVIDIA DGX OS. If you need Windows, this isn't the machine for you.

Q: How much does it cost?

US prices range from USD 6,999 to USD 8,700, and Canadian prices range from CAD 6,579 to CAD 9,499. Shop around because the spread is significant.

Who Should Skip This

Skip this if you need a gaming PC, a general-purpose desktop, or anything running Windows. The gaming score of 22/100 tells you everything. Also skip it if you need lots of ports or rock-solid reliability scores, because this machine is thin on both. And if you're not doing AI development, you're paying a lot for hardware you won't use.

Verdict

Buy this if you're an AI developer who wants to run large models locally without cloud costs or privacy concerns. The 128GB of unified RAM and Grace Blackwell Superchip make it a serious tool for agentic workflows and private inference. It's not for gamers, not for general office work, and not for anyone who wants Windows. But for the right person, it's a quiet, compact, and capable little AI rig.

Usage Scores

종합 (73.8)AI/LLM (38.8)게이밍 (21.6)휴대성 (67.5)크리에이터 (36.2)비즈니스 (67.5)개발 (70.1)Home Office (77.8)Workstation (78.4)
  • 종합73.8
  • AI/LLM38.8
  • 게이밍21.6
  • 휴대성67.5
  • 크리에이터36.2
  • 비즈니스67.5
  • 개발70.1
  • Home Office77.8
  • Workstation78.4

다른 구성4

유사 제품