RUNLOCALAIv38
->Will it run?Best GPUCompareTroubleshootStartLearnPulseModelsHardwareToolsBench
Run check
  1. >
  2. Home
  3. /Hardware
  4. /GMKtec EVO-X3
UNIT · OTHER · DESKTOP
128 GB UNIFIEDworkstation·Reviewed July 2026

GMKtec EVO-X3

— · HARDWARE
GMKtec EVO-X3

No editorial image yet — generic vendor mark shown. Credentials in spec table below.

The EVO-X3 is GMKtec's second-generation Strix Halo mini-PC, launched globally on July 6, 2026 around the same Ryzen AI Max+ 395 (16 Zen 5 cores, Radeon 8060S, up to 126 platform TOPS) that powered the EVO-X2. Its 128 GB of LPDDR5X-8000 is soldered and non-upgradeable, with up to 96 GB allocatable as VRAM; GMKtec claims support for models up to Qwen3-235B. The PS4-sized chassis adds what the EVO-X2 lacked — a PCIe Gen4 x4 OCuLink port for eGPU expansion — alongside USB4 (40 Gbps, 100 W PD), dual M.2 Gen4 slots for up to 16 TB of storage, 2.5GbE, Wi-Fi 7, a 140 W peak performance mode, and GMKtec's pre-installed Claw local-AI launcher. Launch pricing is $3,600 for 128 GB/2 TB and $3,849 for 4 TB — roughly double the EVO-X2's launch price for the same memory and storage.

Released 2026·256 GB/s memory bandwidth
▼ CHECK CURRENT PRICE· 1 retailer
GMKtec EVO-X3
Check on Amazon→

Affiliate disclosure: as an Amazon Associate and partner of other retailers, we earn from qualifying purchases. The verdict on this page is our editorial opinion; affiliate links never influence what we recommend.

RUNLOCALAI SCORE
See full leaderboard →
106/ 1000
DD-tier
Estimated
Throughput
59/ 500
VRAM-fit
0/ 200
Ecosystem
80/ 200
Efficiency
12/ 100

Sub-scores sum to 151 / 1000. Headline = 151 × 0.70 (Estimated-confidence discount) = 106. This is an algorithmic performance-tier score — distinct from, and often lower than, the editorial “Our verdict” below, which weighs value and real-world fit (especially for hardware we haven’t measured yet). How scoring works →

Extrapolated from 256 GB/s bandwidth — 20.5 tok/s estimated. No measured benchmarks yet.

WORKLOAD FIT
Try other hardware →

Plain-English: Doesn't fit modern chat models usefully — vision models won't fit.

7B chat✗
Doesn't fit
14B chat✗
Doesn't fit
32B chat✗
Doesn't fit
70B chat✗
Doesn't fit
Coding agent✗
Doesn't fit
Vision (≤8B VLM)✗
Doesn't fit
Long context (32K)✗
Doesn't fit
✓Comfortable — fits with headroom
~Tight — works, no slack
△Marginal — needs aggressive quant
✗Doesn't fit usefully

Verdicts extrapolated from catalog VRAM + bandwidth + ecosystem flags. Hover any chip for the rationale. Want measured numbers? Submit your own run with runlocalai-bench --submit.

BLK · VERDICT

Our verdict

OP · Eruo Fredoline|VERIFIED JUL 10, 2026
unrated

Positioning

Judge the EVO-X3 on what the extra money buys, because the silicon has not changed: this is the same Ryzen AI Max+ 395 and 128 GB of LPDDR5X-8000 as the EVO-X2, now at $3,600 — double that machine's launch price for the identical memory/storage configuration. That delta is the memory supply crunch showing up on a price tag, not added performance. What is genuinely new is the OCuLink Gen4 x4 port, and it matters: it is the cleanest path in this class to bolt on a discrete GPU for prefill-heavy work, the exact scenario where StorageReview measured this silicon trailing the DGX Spark by up to 8.8x. Against AMD's own Ryzen AI Halo at $3,999, the EVO-X3 is $400 cheaper and adds OCuLink, but gives up 10GbE (2.5GbE here), the first-party dual-OS developer images, and Micro Center-backed support — for most US buyers the Halo is the safer spend. A used M2 Ultra Mac Studio remains the bandwidth play at 800 GB/s versus 256 GB/s; decode speed on large dense models will favor it decisively, though you lose x86, ROCm, and the eGPU option. Buy the EVO-X3 if you specifically want OCuLink expansion or cannot reach a Micro Center. Skip it if you can still find EVO-X2 stock near its original price — it is the same computer — or if $3,600 feels like Spark-adjacent money, since base Spark-class systems sell near $4,000.

BLK · OVERVIEW

Overview

The EVO-X3 is GMKtec's second-generation Strix Halo mini-PC, launched globally on July 6, 2026 around the same Ryzen AI Max+ 395 (16 Zen 5 cores, Radeon 8060S, up to 126 platform TOPS) that powered the EVO-X2. Its 128 GB of LPDDR5X-8000 is soldered and non-upgradeable, with up to 96 GB allocatable as VRAM; GMKtec claims support for models up to Qwen3-235B. The PS4-sized chassis adds what the EVO-X2 lacked — a PCIe Gen4 x4 OCuLink port for eGPU expansion — alongside USB4 (40 Gbps, 100 W PD), dual M.2 Gen4 slots for up to 16 TB of storage, 2.5GbE, Wi-Fi 7, a 140 W peak performance mode, and GMKtec's pre-installed Claw local-AI launcher. Launch pricing is $3,600 for 128 GB/2 TB and $3,849 for 4 TB — roughly double the EVO-X2's launch price for the same memory and storage.

Retailers we'd check:Amazon

Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.

BLK · SPECS

Specs

System RAM (typical)128 GB
Power draw (peak)140 W
Released2026
MSRP$3600
Backends
ROCm
Vulkan

Models that fit

Open-weight models small enough to run on GMKtec EVO-X3 with usable context.

all-MiniLM-L6-v2
0.022B · other
FLUX.1 [dev]
12B · other
Qwen 3 0.6B
0.6B · qwen
BGE Large EN v1.5
0.335B · other
Nomic Embed Text v1.5
0.137B · other
Kokoro 82M
0.082B · other
Llama 3.1 8B Instruct
8B · llama
Qwen 3 30B-A3B
30B · qwen

Frequently asked

Does GMKtec EVO-X3 support CUDA?

No — GMKtec EVO-X3 is an AMD card. Use ROCm (Linux) or the Vulkan backend in llama.cpp instead. CUDA-only tools won't work.

Where next?

Buyer guides
  • Best GPU for local AI →
  • Best laptop for local AI →
  • Best Mac for local AI →
  • Best used GPU for local AI →
Troubleshooting
  • CUDA out of memory →
  • Ollama running slowly →
  • ROCm not detected →
  • Model keeps crashing →

Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify hardware specifications.

RUNLOCALAI

Independently operated catalog for local-AI hardware and software. Hand-written verdicts. Source-cited claims. Reproducible commands when we have them.

OP·Eruo Fredoline
DIR
  • Models
  • Hardware
  • Tools
  • Benchmarks
TOOLS
  • Will it run?
  • Compare hardware
  • Cost vs cloud
  • Choose my GPU
  • Prompting kits
  • Quick answers
REF
  • All buyer guides
  • Learn local AI
  • Methodology
  • Glossary
  • Errors KB
  • Trust
EDITOR
  • About
  • Author
  • How we make money
  • Editorial policy
  • Contact
LEGAL
  • Privacy
  • Terms
  • Sitemap
MAIL · MONTHLY DIGEST
Get monthly local AI changes
Monthly recap. No spam.
DISCLOSURE

Some links on this site are affiliate links (Amazon Associates and other first-class retailers). When you buy through them, we earn a small commission at no extra cost to you. Affiliate links do not influence our verdicts — there are cards we rate highly that we don't have affiliate relationships with, and cards that sell well that we refuse to recommend. Read more →

© 2026 runlocalai.coIndependently operated
RUNLOCALAI · v38
Compare alternatives

Hardware worth comparing

The closest alternatives by price, memory bandwidth, and form factor, plus a step up and down — so you can frame the buying decision against real options.

Closest matches
Similar price, bandwidth & form factor
  • AMD Ryzen AI Halo
    amd · 256 GB/s
    unrated
  • ASUS Ascent GX10 (NVIDIA GB10)
    nvidia · 273 GB/s
    8.1/10
  • NVIDIA DGX Spark (Project Digits)
    nvidia · 273 GB/s
    10.0/10
  • Framework Desktop (Ryzen AI Max+ 395)
    amd · 256 GB/s
    8.2/10
  • GMKtec EVO-X2 (Ryzen AI Max+ 395)
    amd · 256 GB/s
    8.0/10
  • Intel Arc Pro B60 24GB
    intel · 24 GB VRAM
    7.6/10
Step up
More capable — more memory or a higher tier
  • NVIDIA RTX 5000 PRO Blackwell 48GB
    nvidia · 48 GB VRAM
    8.5/10
  • NVIDIA A40
    nvidia · 48 GB VRAM
    9.7/10
  • NVIDIA RTX 6000 Ada Generation
    nvidia · 48 GB VRAM
    10.0/10
Step down
Lighter — cheaper or more constrained
  • Framework Desktop (Ryzen AI Max+ 395)
    amd · 256 GB/s
    8.2/10
  • NVIDIA RTX PRO 4000 Blackwell
    nvidia · 24 GB VRAM
    7.3/10
  • Intel Arc Pro B60 24GB
    intel · 24 GB VRAM
    7.6/10