COMPARISON DESK
Compare runs without hiding the mismatches.
A faster number means little if the hardware, task set, or concurrency changed.
ComparableWarning: Runtime differsWarning: Quant differsWarning: Reasoning effort differs
| Run | Score | Decode | Runtime | Configuration | Evidence |
|---|---|---|---|---|---|
| Qwen3.6 NVFP4qwen36-nvfp4-agent-2026-08-28 | 7/9 | 61.5 tok/s | FreeToken | NVFP4 · 32,768 ctx · c1 | Documented local result |
| Flash-Next Q3_K_XLflash-next-q3kxl-agent-2026-08-28 | 9/9 | 14.5 tok/s | llama.cpp | Q3_K_XL · 32,768 ctx · c1 | Documented local result |