Machine
RTX 5060 Ti + RTX 5090 · Ryzen 9 9950X #793f5ff2
4 runs of 3 models since Nov 4, 2026. Last run Nov 18, 2026.
4
runs
3
models
3
configs
1
configs run 2+ times
Hardware
- GPU
- NVIDIA GeForce RTX 5060 Ti + NVIDIA GeForce RTX 5090 47 GB
- CPU
- AMD Ryzen 9 9950X 16-Core Processor 16 cores
- RAM
- 63 GB
- OS
- Windows 11 Home 25H2
- Tools
- llama.cpp 3 · Strata 1
- Client
- 0.4.98
Best config per model
sort
Runs with the same model, quant, tool, context, thinking and KV cache count as one config. Numbers are the median of its runs, and each dot is one run on the 0–100 quality scale. ★ marks the best config with at least 2 rated runs.
configrunsqualityeach runspeed
qwen3.8-flash-next-iq3_s
1 run · 1 config
Qwen3.8-Flash-Next NVIDIA-NVFP4
1 run · 1 config
qwen3.8-flash-next-unsloth-ud-iq4_xs
2 runs · 1 config
★ bestllama.cpp262k ctxthinking defaulteffort xhigh
×2 runs
84.780.3–89.1
varies
52.4 tok/s
49.3–55.5