AMD Radeon RX 6900 XT × Qwen2.5 14B Instruct
Qwen2.5 14B Instruct on AMD Radeon RX 6900 XT
24.5best gen tok/s
612best 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_K - Medium | 24.5 | 612 | LocalScore ↗ | 512+128 | 2,398 ms | llamafile 0.9.2 |
| Q4_K - Medium | 24.2 | 574 | LocalScore ↗ | 512+128 | 2,565 ms | llamafile 0.9.2 |
| Q4_K - Medium | 22.2 | 593 | LocalScore ↗ | 512+128 | 2,463 ms | llamafile 0.9.2 |
| Q4_K - Medium | 3.0 | 496 | LocalScore ↗ | 512+128 | 3,113 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 AMD Radeon RX 6900 XT 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.