RTX 4000 SFF Ada Generation × Qwen2.5 14B Instruct
Qwen2.5 14B Instruct on RTX 4000 SFF Ada Generation
24.9best gen tok/s
1,066best prompt tok/s
640max context tested
2results
Measured · all numbers below come from the linked sources
llama.cpp
| Quant | Gen tok/s | Prompt tok/s | Source | Context tested | TTFT | Build |
|---|---|---|---|---|---|---|
| Q4_K - Medium | 24.9 | 1,037 | LocalScore ↗ | 512+128 | 1,343 ms | llamafile 0.9.2 |
| Q4_K - Medium | 24.5 | 1,066 | LocalScore ↗ | 512+128 | 1,265 ms | llamafile 0.9.2 |
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.
What else can RTX 4000 SFF Ada Generation run? →
Includes results from LocalScore, a Mozilla Builders project by cjpais, mirrored in the cn0303/local-llm-speed-benchmarks dataset. LocalScore runs use llamafile, which is built on llama.cpp.