Qwen2.5 14B Instruct › NVIDIA DGX Spark (GB10)
Qwen2.5 14B Instruct on the NVIDIA DGX Spark (GB10)
Best measured: 18.3 tokens/s generation and 1,094 tokens/s prompt processing (Q4_K_M, llamafile).
For comparison, our size-scaling estimate for this combination is 26.5 tok/s.
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| Q4_K_M | llamafile | 18.4 | 1,119 | LocalScore avg | LocalScore rank 65/418 | LocalScore (localscore.ai) |
| Q4_K_M | llamafile | 18.3 | 1,069 | LocalScore avg | LocalScore rank 68/418 | LocalScore (localscore.ai) |
tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128.
No card yet?
Try it on a rented GPU (RunPod)Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.
All results for the NVIDIA DGX Spark (GB10) · Qwen2.5 14B Instruct on other hardware