Machine
Ryzen 9 9950X3D2 #588ac2e8
9 runs of 4 models since Nov 22, 2026. Last run Nov 22, 2026.
9
runs
4
models
7
configs
2
configs run 2+ times
Hardware
- CPU
- AMD Ryzen 9 9950X3D2 16-Core Processor 16 cores
- RAM
- 93 GB
- OS
- Windows 11 Pro 26H2
- Tools
- llama.cpp 9
- Client
- 0.4.99
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-Q4_K_M
2 runs · 2 configs
Qwen3.8-Flash-Next-heretic-2-IQ4_XS-MTP-00001-of-00006
1 run · 1 config
Muse-Glimmer-30B-KQuant-17GB-Q4_K_M
2 runs · 2 configs
Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated.i1-Q4_K_M
4 runs · 2 configs
★ bestQ4_K_Mllama.cpp64k ctxthinking default
×2 runs
61.357.0–65.6
varies
110 tok/s
110–110
Compare with another machine
Run history
Nov 22, 2026
04:55Qwen3.8-Flash-Next-heretic-2-IQ4_XS-MTP-00001-of-00006llama.cpp30.8 tok/s87.204:14Muse-Glimmer-30B-KQuant-17GB-Q4_K_Mllama.cpp65.7 tok/s82.104:01Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated.i1-Q4_K_Mllama.cpp110 tok/s57.903:48Qwen3.8-27B-UD-Q4_K_Mllama.cpp47.6 tok/s87.203:03Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated.i1-Q4_K_Mllama.cpp110 tok/s58.402:58Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated.i1-Q4_K_Mllama.cpp110 tok/s65.602:43Muse-Glimmer-30B-KQuant-17GB-Q4_K_Mllama.cpp66.0 tok/s85.402:35Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated.i1-Q4_K_Mllama.cpp110 tok/s57.001:51Qwen3.8-27B-UD-Q4_K_Mllama.cpp46.5 tok/s86.8