Machine
RTX 4090 · Ryzen 9 7950X3D #85582241
6 runs of 4 models since Nov 2, 2026. Last run Nov 7, 2026.
6
runs
4
models
4
configs
1
configs run 2+ times
Hardware
- GPU
- NVIDIA GeForce RTX 4090 24 GB
- CPU
- AMD Ryzen 9 7950X3D 16-Core Processor 16 cores
- RAM
- 63 GB
- OS
- Windows 11 Education Insider Preview 26H1
- Tools
- llama.cpp 6
- 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
Qwen3.8-27B-UD-IQ4_XS
1 run · 1 config
Qwen3.8-27B-UD-IQ3_S
3 runs · 1 config
★ bestllama.cpp128k ctxthinking oneffort xhigh
×3 runs
88.387.7–88.6
steady1 slow run
94.2 tok/s
58.2–99.8
Qwen3.8-27B-UD-Q4_K_M
1 run · 1 config
Qwen3.8-27B-TurboFCFusion-735-882-Here-Uncen-NEO-CODER-MAX-MTP-Q4_K_S
1 run · 1 config