NVIDIA CMP 170HX 8GB vs NVIDIA DGX Spark for local LLMs
1 report on the NVIDIA CMP 170HX 8GB and 88 on the NVIDIA DGX Spark
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 5.5x higher on the NVIDIA CMP 170HX 8GB (1,493 GB/s on the NVIDIA CMP 170HX 8GB, 273 on the NVIDIA DGX Spark). The estimates below compare them on the models people run most.
- Memory
- 8GB
- Memory bandwidth
- 1,493 GB/s
- FP16 compute
- 6.3 TFLOPS
- Reports
- 1
- Memory
- 128GB unified
- Memory bandwidth
- 273 GB/s
- FP16 compute
- 125 TFLOPS
- Reports
- 88
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.
| Model | NVIDIA CMP 170HX 8GB | NVIDIA DGX Spark |
|---|---|---|
| Qwen3.8 27B | 2.1 t/sQ6_K, layers on CPU | 4.0 t/sFP16 |
| Qwen3.8 125B · 6B active | doesn't fit at these settings | 14.4 t/sQ6_K |
| Qwen3.6 35B · 3B active | 18.3 t/sQ5_K_M, experts in RAM | 14.5 t/sFP16 |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | 38.3 t/sQ2_K |
| Qwen3.6 27B | 2.1 t/sQ6_K, layers on CPU | 4.0 t/sFP16 |
| Gemma 4 26B · 4B active | 11.6 t/sQ6_K, experts in RAM | 10.8 t/sFP16 |
Change the context, RAM or model in the calculator: NVIDIA CMP 170HX 8GB · NVIDIA DGX Spark
Every report on each card, with its source: NVIDIA CMP 170HX 8GB · NVIDIA DGX Spark