How we got this number
Each generated token reads the whole model from memory once: 7.4 GB. The NVIDIA DGX Spark (128GB) moves 273 GB/s, so the ceiling is 37 tokens/sec. Real machines reach a fraction of that ceiling; we use 70%, an assumption until we get measured runs for this kind of machine. Send us your real numbers and we'll replace the estimate.
Faster machines for Llama 2 13B
Other models on the NVIDIA DGX Spark (128GB)
| lfm 2.5 8B (A1B) | Fast | 120+ tok/s |
| minicpm v4.6 1B | Fast | 119 tok/s |
| Gemma 2 2B | Fast | 119 tok/s |
| codeGemma 2B | Fast | 119 tok/s |
| Gemma 2B | Fast | 112 tok/s |
| smollm 2 1.7B | Fast | 106 tok/s |