readygpu

NVIDIA DGX Spark 128GB vs NVIDIA GeForce RTX 4090

For running models locally, priced in the UK. At 32k context the first runs 76 models fully in memory, the second 54.

In short

The NVIDIA DGX Spark 128GB has more memory for models: 128 GB against 24 GB. At 32k context the NVIDIA DGX Spark 128GB runs 76 of the models we track fully in memory and the NVIDIA GeForce RTX 4090 54; only the NVIDIA DGX Spark 128GB holds DeepSeek V4 Flash. On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 3.7× as fast as the NVIDIA DGX Spark 128GB (estimated, median across models). In the UK the NVIDIA DGX Spark 128GB costs £7,000 at eBay and the NVIDIA GeForce RTX 4090 £6,015 at Amazon: the NVIDIA GeForce RTX 4090 is £985 cheaper.

Updated . Speeds are estimates; how they are worked out.

NVIDIA DGX Spark 128GB

£7,000

eBay

Add to build
NVIDIA GeForce RTX 4090

£6,015

Amazon

Add to build

Specifications

SpecNVIDIA DGX Spark 128GBNVIDIA GeForce RTX 4090
ChipGB10 Grace Blackwell–
Memory bus256-bit384-bit
Memory bandwidth273 GB/s1,008 GB/s
FP16 compute62 TFLOPS165 TFLOPS
SoftwareCUDACUDA
Unified memory128 GB–
GPU can use by default119 GB–
GPU can use, tuned122 GB–
Operating systemdgx-os–
Processor20-core Arm (GB10)–
Storage4000 GB–
GPU–GeForce RTX 4090
Memory–24 GB GDDR6X
Board power–450 W
PCIe generation–Gen 4
PCIe lanes–x16

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelNVIDIA DGX Spark 128GBNVIDIA GeForce RTX 4090
DeepSeek V4 Flash11tok/sUD-Q2_K_XLwon't run
GLM 5.3 Flash11tok/sUD-Q2_K_XLwon't run
Qwen 3.8 Flash Next2tok/sQ4_K_Mwon't run
MiniMax M2.78tok/sUD-Q3_K_Mwon't run
Qwen 3.8 27B5tok/sQ8_027tok/sQ5_K_M
Qwen 3.5 122B-A10B11tok/sQ6_Koffload
Qwen 3.6 27B6tok/sQ8_029tok/sQ5_K_M
gpt-oss 120B23tok/sMXFP4won't run
Qwen3 Coder Next20tok/sQ8_0offload
Nemotron 3 Super 120B-A12B9tok/sUD-Q6_Kwon't run
Gemma 4 31B5tok/sQ8_029tok/sQ4_K_M
Mistral Small 4 119B15tok/sUD-Q6_Koffload
Qwen 3.6 35B-A3B27tok/sQ8_0145tok/sQ4_K_M
Qwen 3.5 27B6tok/sQ8_029tok/sQ5_K_M
Gemma 4 26B-A4B20tok/sQ8_084tok/sUD-Q6_K
GLM 4.5 Air4tok/sQ8_0offload
Qwen 3.5 35B-A3B27tok/sQ8_0146tok/sQ4_K_M
Qwen3 Next 80B-A3B25tok/sQ8_0offload
GLM 4.7 Flash14tok/sQ8_070tok/sQ5_K_M
Devstral Small 2 24B6tok/sQ8_028tok/sQ5_K_M
Nemotron 3.5 Lightning 30B-A3B28tok/sUD-Q6_K_XL153tok/sMXFP4
gpt-oss 20B31tok/sMXFP4116tok/sMXFP4
Kimi Linear 48B-A3B29tok/sQ8_0offload
Qwen3 Coder 30B-A3B16tok/sQ8_074tok/sQ4_K_M
Qwen3 32B4tok/sQ8_0offload
Gemma 4 12B13tok/sQ8_046tok/sQ8_0
Qwen 3.5 9B16tok/sQ8_059tok/sQ8_0
Mistral Small 3.2 24B6tok/sQ8_028tok/sQ5_K_M
Gemma 3 27B5tok/sQ8_028tok/sQ5_K_M
DeepSeek R1 Distill Qwen 32B4tok/sQ8_0offload
Llama 3.3 70B2tok/sQ8_0won't run
Ministral 3 14B9tok/sQ8_032tok/sQ8_0
Qwen3 14B8tok/sQ8_030tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B13tok/sQ8_046tok/sQ8_0
Qwen3 8B13tok/sQ8_046tok/sQ8_0
Phi-4 14B8tok/sQ8_028tok/sQ8_0
Qwen 3.5 4B30tok/sQ8_0112tok/sQ8_0
Ministral 3 8B13tok/sQ8_046tok/sQ8_0
Granite 4.2 8B12tok/sQ8_043tok/sQ8_0
Gemma 4 E4B19tok/sQ8_068tok/sQ8_0
LFM2.5 8B-A1B38tok/sQ8_0140tok/sQ8_0
Llama 3.1 8B16tok/sQ6_K57tok/sQ6_K
Qwen2.5 0.5B181tok/sQ8_0669tok/sQ8_0
Qwen3.5 0.8B137tok/sQ8_0507tok/sQ8_0
Qwen3.5 2B91tok/sQ5_K_M337tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B60tok/sQ8_0221tok/sQ8_0
MiniCPM5 2B41tok/sQ8_0153tok/sQ8_0
Spark X2.5 4B39tok/smlx-4bit145tok/smlx-4bit
LFM2.5 2.6B35tok/sQ8_0128tok/sQ8_0
LFM2.5 VL 3B35tok/sQ8_0128tok/sQ8_0
Mellum2.1 12B A2.5B Thinking32tok/sQ8_0118tok/sQ8_0
FrogNano 4B 260930tok/sQ8_0109tok/sQ8_0
granite vision 4.1 4b27tok/sQ8_099tok/sQ8_0
granite 4.2 3b26tok/sQ8_095tok/sQ8_0
Kimi VL A3B Thinking 250625tok/sQ8_091tok/sQ8_0
Kimi VL A3B24tok/smlx-8bit88tok/smlx-8bit
LFM2.5 24B A2B21tok/sQ8_084tok/sQ6_K
GLM 4 9B 041418tok/smlx-5bit65tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B17tok/sQ8_063tok/sQ8_0
GLM 4.6V Flash15tok/sQ8_055tok/sQ8_0
DeepSeek R1 Distill Llama 8B13tok/sQ8_049tok/sQ8_0
GLM 4.1V 9B Thinking13tok/smlx-8bit48tok/smlx-8bit
Ministral 3 8B Reasoning 251213tok/sQ8_046tok/sQ8_0
DeepSeek R1 Distill Qwen 14B8tok/sQ8_028tok/sQ8_0
Qwen3.8 27B OBLITERATED7tok/smlx-4bit27tok/smlx-4bit
Devstral Small 25076tok/sQ8_028tok/sQ5_K_M
Magistral Small 25066tok/sQ8_028tok/sQ5_K_M
Magistral Small 25076tok/sQ8_028tok/sQ5_K_M
Magistral Small 25096tok/sQ8_028tok/sQ5_K_M
GLM 4.5V5tok/sQ8_0won't run
GLM 4.6V5tok/sQ8_0won't run
granite 4.2 30b4tok/sQ8_0offload
DeepSeek R1 Distill Llama 70B2tok/sQ8_0won't run
Kimi Dev 72B2tok/sQ8_0won't run
Devstral 2 123B Instruct 25122tok/sQ6_Kwon't run
Mistral Medium 3.5 128B2tok/sQ6_Kwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3 0324won't runwon't run
DeepSeek V3.1won't runwon't run
DeepSeek V3.1 Terminuswon't runwon't run
DeepSeek V3.2won't runwon't run
DeepSeek V4 Flash 0731won't runwon't run
DeepSeek V4 Flash Vision Expwon't runwon't run
GLM 4.6won't runwon't run
GLM 4.7won't runwon't run
GLM 5won't runwon't run
GLM 5.1won't runwon't run
GLM 5.2won't runwon't run
GLM 5.3won't runwon't run
Kimi K2won't runwon't run
Kimi K2 Instruct 0905won't runwon't run
Kimi K2 Thinkingwon't runwon't run
Kimi K2.5won't runwon't run
MiniMax M2won't runwon't run
MiniMax M2.1won't runwon't run
MiniMax M2.5won't runwon't run
MiniMax M3won't runwon't run

Questions

Which is better for local AI, the NVIDIA DGX Spark 128GB or the NVIDIA GeForce RTX 4090?
Choose the NVIDIA DGX Spark 128GB to run bigger models. Choose the NVIDIA GeForce RTX 4090 for speed. Choose the NVIDIA GeForce RTX 4090 to spend less.
Which can run bigger models, the NVIDIA DGX Spark 128GB or the NVIDIA GeForce RTX 4090?
At 32k context the NVIDIA DGX Spark 128GB runs 76 of the models we track fully in memory and the NVIDIA GeForce RTX 4090 54; only the NVIDIA DGX Spark 128GB holds DeepSeek V4 Flash.
Which is faster for local AI, the NVIDIA DGX Spark 128GB or the NVIDIA GeForce RTX 4090?
On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 3.7× as fast as the NVIDIA DGX Spark 128GB (estimated, median across models).
What is the price difference between the NVIDIA DGX Spark 128GB and the NVIDIA GeForce RTX 4090?
In the UK the NVIDIA DGX Spark 128GB costs £7,000 at eBay and the NVIDIA GeForce RTX 4090 £6,015 at Amazon: the NVIDIA GeForce RTX 4090 is £985 cheaper.

More comparisons