gpu4.pics
data via LocalScore
AMD Radeon RX 6650 XT
llamafile 0.9.2
8 GB VRAM AMD
running Qwen2.5.1 Coder 7B Instruct· Q4_K_M· 7.6B
26.8
tokens / sec
180 · Modest
Generation
26.8tok/s
higher is better
Prompt / prefill
566tok/s
higher is better
Time to first token
2.58 s
lower is better
Composite score
180
Model size
7.6B
Runtime
llamafile
All benchmarks on this rig
| Model | Quant | Gen | Prompt | TTFT | Power | Perf/W | $/Mtok | Score |
|---|---|---|---|---|---|---|---|---|
| Qwen2.5.1 Coder 7B Instruct7.6B | Q4_K_M | 26.8 | 566 | 2.58 s | — | — | — | 180 |