llamaperf

NVIDIA DGX Spark vs AMD RX 5700 XT 8GB for local LLMs

89 reports on the NVIDIA DGX Spark and 1 on the AMD RX 5700 XT 8GB

Which is faster for local LLMs?

Nobody has reported the same model at the same quant on both cards yet, so there's no measured head-to-head. Memory bandwidth, which caps how fast a card can write, is 1.6x higher on the AMD RX 5700 XT 8GB (273 GB/s on the NVIDIA DGX Spark, 448 on the AMD RX 5700 XT 8GB). The estimates below compare them on the models people run most.

NVIDIA DGX Spark
Memory
128GB unified
Memory bandwidth
273 GB/s
FP16 compute
125 TFLOPS
Reports
89
AMD RX 5700 XT 8GB
Memory
8GB
Memory bandwidth
448 GB/s
FP16 compute
19.5 TFLOPS
Reports
1

Estimated by the calculator

The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape.

ModelNVIDIA DGX SparkAMD RX 5700 XT 8GB
Qwen3.8 27B4.0 t/sFP161.8 t/sQ6_K, layers on CPU
Qwen3.8 125B · 6B active14.4 t/sQ6_Kdoesn't fit at these settings
Qwen3.6 35B · 3B active14.5 t/sFP1616.1 t/sQ5_K_M, experts in RAM
DeepSeek V4 Flash 284B · 13B active38.3 t/sQ2_Kdoesn't fit at these settings
Qwen3.6 27B4.0 t/sFP161.8 t/sQ6_K, layers on CPU
Gemma 4 26B · 4B active10.8 t/sFP1610.2 t/sQ6_K, experts in RAM

Change the context, RAM or model in the calculator: NVIDIA DGX Spark · AMD RX 5700 XT 8GB

Every report on each card, with its source: NVIDIA DGX Spark · AMD RX 5700 XT 8GB

More comparisons