Llama 2 7B on DGX Spark
57.2best gen tok/s
3,661best prompt tok/s
640max context tested
4results
Measured · all numbers below come from the linked sources
llama.cpp
| Quant | Gen tok/s | Prompt tok/s | Source | Context tested | TTFT | Build |
|---|---|---|---|---|---|---|
| Q4_0 | 57.2 | 3,062 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @5acd455 |
| Q4_0 | 56.7 | 3,661 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @5acd455 |
| Q4_0 | 53.6 | 3,280 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @b9da444 — coopmat2 |
| Q4_0 | 52.3 | 2,738 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @b9da444 — coopmat2 |
Ollama
No measured Ollama results yet. Submit yours →
MLX
No measured MLX results yet. Submit yours →
"Context tested" is the largest prompt+generation length actually benchmarked (tokens), not a memory-fit estimate.