ASUS Ascent GX10 vs NVIDIA DGX Spark for local AI
NVIDIA DGX Spark should generate tokens faster (more memory bandwidth). Both fit the same model sizes. ASUS Ascent GX10 is cheaper right now by $951.00.
Specs
| ASUS Ascent GX10 | NVIDIA DGX Spark | |
|---|---|---|
| Stock | Out of stock | Available |
| Lowest price | $5,999 | $6,950 |
| Memory | 128 GB LPDDR5x unified | 128 GB LPDDR5x unified |
| Usable for models (est.) | ~109 GB | ~109 GB |
| Bandwidth | 273 GB/s | 273 GB/s |
| Software | CUDA | CUDA |
| Price per GB | $46.87 | $54.30 |
Model by model (4-bit)
| Model | ASUS Ascent GX10 | NVIDIA DGX Spark |
|---|---|---|
| Qwen3.5 9B | Fits~30 tok/s | Fits~30 tok/s |
| Gemma 4 12B | Fits~23 tok/s | Fits~23 tok/s |
| GPT-OSS 20B | Fits~86 tok/s | Fits~86 tok/s |
| Gemma 4 26B-A4B | Fits~71 tok/s | Fits~71 tok/s |
| Qwen3.8 27B | Fits~10 tok/s | Fits~10 tok/s |
| GLM-4.7-Flash 30B-A3B | Fits~90 tok/s | Fits~90 tok/s |
| Gemma 4 31B | Fits~9 tok/s | Fits~9 tok/s |
| Qwen3.6 35B-A3B | Fits~90 tok/s | Fits~90 tok/s |
| Llama 3.3 70B | Fits~4 tok/s | Fits~4 tok/s |
| Qwen3-Next 80B-A3B | Fits~90 tok/s | Fits~90 tok/s |
| Llama 4 Scout 109B-A17B | Fits~16 tok/s | Fits~16 tok/s |
| GPT-OSS 120B | Fits~60 tok/s | Fits~60 tok/s |
| Nemotron 3 Super 120B-A12B | Fits~23 tok/s | Fits~23 tok/s |
| Qwen3.5 122B-A10B | Fits~27 tok/s | Fits~27 tok/s |
| Qwen3 235B-A22B | Won't fit | Won't fit |
| Muse Spark Open 816B-A42B | Won't fit | Won't fit |