Versus Engine
GMKtec EVO-X2 (128GB) vs Radeon RX 9070 GRE
Specs, price and the one trade-off that actually decides it — GMKtec EVO-X2 (128GB) against Radeon RX 9070 GRE, side by side.
Cheaper to start
Tie
Both start at a similar price.
Rated higher
GMKtec EVO-X2 (128GB)
8.0 vs 0.0 on our design scale.
Standout
GMKtec EVO-X2 (128GB)
128GB massive memory pool
GMKtec EVO-X2 (128GB)
Enthusiast-tier mini workstation: AMD Ryzen AI Max+ 395 with 128 GB of unified memory. Runs Llama 3.3 70B and Mistral Small 3.1 at roughly 15-25 tokens/s — 70B-class models without a discrete GPU, in a small-form-factor chassis.
Where it wins, where it doesn't
Pros
- 128GB massive memory pool
- Ultra-compact form factor
- Dual NVMe storage slots
Cons
- APU graphics bottleneck token generation speed
- Audible fan noise under sustained load
Specifications
- memory_speed8000MHz
- achievability, frequency, and sustainability—
- AI Accelerators—
- ai_npuBased on AMD XDNA 2 architecture; AI TOPS performance varies by application, model, and settings
- Architecture—
- audio3.5mm headphone / audio jack ×1
- Bandwidth—
- Base Clock—
- bluetoothBluetooth 5.4
- Boost Clock—
- Bus Type—
- chipAMD Ryzen AI Max+ 395
- Chip Package—
- Compute Units—
- Connectivity—
- Density—
- Die Size—
- dimensionsApproximately 186 × 77 × 192 mm
- DirectX—
- Display Core Next—
- display_outputHDMI 2.1 ×1, DisplayPort 1.4 ×1, USB4 ×2, supports up to 8K@60Hz display output, actual output depends on display, cables, and settings
- displayportDisplayPort 1.4 ×1, up to 8K@60Hz
- example_modelsLlama 3.3 70B, Mistral Small 3.1
- Features—
- Foundry—
- FP16—
- FP32—
- FP64—
- Game Clock—
- Game Frequency—
- Generation—
- GPU Memory—
- GPU Name—
- GPU Variant—
- Graphics Card—
- Graphics/Compute—
- hdmiHDMI 2.1 ×1, up to 8K@60Hz
- Infinity Cache—
- INT4—
- INT8—
- JPEG Engine—
- L0 Cache—
- L2 Cache—
- L3 Cache—
- Matrix Cores—
- Max Memory Size—
- memory128GB LPDDR5X 8000MHz
- Memory Bandwidth—
- Memory Bus—
- Memory Clock—
- Memory Interface—
- Memory Size—
- Memory Type—
- memory_capacity128GB LPDDR5X
- memory_formOnboard LPDDR5X, typically not user-upgradeable
- Minimum PSU Recommendation—
- nameGMKtec EVO-X2 AMD Ryzen AI Max+ 395 128GB RAM + 2TB SSD Windows 11 Pro Mini PC 2 years warranty (CS-GEVOX2)
- OpenCL—
- OpenGL—
- operating_systemWindows 11 Pro
- Outputs—
- Package Size—
- Peak 4-bit Precision (INT4 Matrix) Performance—
- Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3)—
- Peak 8-bit Precision (INT8 Matrix) Performance—
- Peak Half Precision (FP16 Matrix) Performance—
- Peak Half Precision (FP16 Vector) Performance—
- Peak Pixel Fill-Rate—
- Peak Single Precision (FP32 Vector) Performance—
- Pixel Rate—
- Power Connectors—
- Predecessor—
- preloaded_storage2TB M.2 2280 SSD
- Process Size—
- Process Type—
- processorAMD Ryzen™ AI Max+ 395, 16 cores / 32 threads, up to 5.1GHz
- Radiance Caching—
- Radiance Display Engine—
- Ray Accelerators—
- Ray Regeneration—
- Ray Tracing—
- ROPs—
- RT Cores—
- sd_card_readerSome official / retail data indicates SD Card Reader, actual shipping version varies
- Shader ISA—
- Shader Model—
- skuCS-GEVOX2
- Slot Width—
- storage2TB M.2 2280 SSD; up to 2 M.2 2280 slots for expansion, actual capacity depends on SKU and hardware support
- storage_expansionUp to 2 M.2 2280 slots; expansion limit depends on actual unit and manufacturer data
- Stream Processors—
- Successor—
- Suggested PSU—
- Supported Rendering Format—
- TDP—
- tierenthusiast
- TMUs—
- tokens_per_second15-25
- TPP—
- TPP Density—
- Transistor Count—
- Transistors—
- Typical Board Power (Desktop)—
- unified_memory_gb128
- usb_aUSB 3.2 ×3, USB 2.0 ×2
- Video Core Next—
- Vulkan—
- warranty2 years
- wired_network2.5G LAN ×1
- wireless_networkWi-Fi 7
Radeon RX 9070 GRE
High-performance Radeon RX 9070 GRE graphics card with 12 GB GDDR6 memory
Where it wins, where it doesn't
Pros
- High peak performance with up to 274 TFLOPs in 8-bit precision
- Support for advanced video formats and AI accelerators
- Wide range of connectivity options including DisplayPort 2.1a and HDMI 2.1b
Cons
- High power consumption of 220W typical board power
- Minimum PSU recommendation of 650W
- Limited upgradeability due to soldered memory
Specifications
- memory_speedUp to 18 Gbps
- achievability, frequency, and sustainabilitywill vary based on several factors, including but not limited to: thermal conditions and variation in applications and workloads.
- AI Accelerators96
- ai_npu—
- ArchitectureRDNA 4
- audio—
- Bandwidth432.0 GB/s
- Base Clock1420 MHz
- bluetooth—
- Boost Clock2790 MHz
- Bus TypePCIe® 5.0 x16
- chip—
- Chip PackageHFCBGA-2069
- Compute Units4th Gen
- ConnectivityDisplayPort™ 2.1a, HDMI™ 2.1b
- Density151.0M / mm²
- Die Size357 mm²
- dimensions—
- DirectX12 Ultimate (12_2)
- Display Core Next4.0.1
- display_output—
- displayport—
- example_models—
- FeaturesAMD Software Adrenalin Edition, AMD FSR™ Redstone, AMD Fluid Motion Frames, AMD Radeon™ Super Resolution, AMD Smart Access Memory, AMD Smart Access Video, AMD Radeon™ Boost, AMD Radeon™ Anti-Lag 2, AMD Radeon™ Image Sharpening 2, AMD Enhanced Sync Technology, AMD FreeSync™ Technology, AMD Radeon™ Chill, AMD Virtual Super Resolution
- FoundryTSMC
- FP1634.28 TFLOPS (1:1)
- FP3234.28 TFLOPS
- FP641.071 TFLOPS (1:32)
- Game Clock2220 MHz
- Game FrequencyUp to 2790 MHz
- GenerationNavi IV
- GPU Memory48 MB
- GPU NameNavi 48
- GPU VariantNavi 48 XL
- Graphics CardRX 9000
- Graphics/Compute12.0.1
- hdmi—
- Infinity Cache3rd Gen
- INT4548.5 TOPS
- INT8274.3 TOPS
- JPEG Engine5.0.0
- L0 Cache32 KB per CU
- L2 Cache6 MB
- L3 Cache48 MB
- Matrix Cores96
- Max Memory Size12 GB
- memory—
- Memory BandwidthUp to 432 GB/s
- Memory Bus192 bit
- Memory Clock2250 MHz
- Memory Interface192-bit
- Memory Size12 GB
- Memory TypeGDDR6
- memory_capacity—
- memory_form—
- Minimum PSU Recommendation650W
- name—
- OpenCL2.0
- OpenGL4.6
- operating_system—
- Outputs1x HDMI 2.1b, 3x DisplayPort 2.1a
- Package Size45 mm × 35 mm
- Peak 4-bit Precision (INT4 Matrix) Performance549 TOPs
- Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3)274 TFLOPs
- Peak 8-bit Precision (INT8 Matrix) Performance274 TOPs
- Peak Half Precision (FP16 Matrix) Performance137 TFLOPs
- Peak Half Precision (FP16 Vector) Performance34.3 TFLOPs
- Peak Pixel Fill-RateUp to 267.8 GP/s
- Peak Single Precision (FP32 Vector) Performance34.3 TFLOPs
- Pixel Rate267.8 GPixel/s
- Power Connectors2x 8-pin
- PredecessorNavi III
- preloaded_storage—
- Process Size4 nm
- Process TypeN4P FinFET
- processor—
- Radiance CachingFSR 0.9.0 (Technical Preview)
- Radiance Display Engine2nd Gen
- Ray Accelerators48
- Ray RegenerationFSR 1.2.1
- Ray Tracing3rd Gen
- ROPs96
- RT Cores48
- sd_card_reader—
- Shader ISAGFX12.0 (gfx1201)
- Shader Model6.9
- sku—
- Slot WidthDual-slot
- storage—
- storage_expansion—
- Stream Processors3072
- SuccessorAlpha Trion
- Suggested PSU550 W
- Supported Rendering FormatHDMI™ 4K Support: Yes, 4K H264 Decode: Yes, 4K H264 Encode: Yes, H265/HEVC Decode: Yes, H265/HEVC Encode: Yes, AV1 Decode: Yes, AV1 Encode: Yes
- TDP220 W
- tier—
- TMUs192
- tokens_per_second—
- TPP2194
- TPP Density6.38
- Transistor Count53.9 B
- Transistors53.9 billion
- Typical Board Power (Desktop)220W
- unified_memory_gb—
- usb_a—
- Video Core Next5.0.0
- Vulkan1.4
- warranty—
- wired_network—
- wireless_network—