Machine
RTX 5090 · Ryzen 9 7900X #a7375cec
20 runs of 15 models since Aug 25, 2026. Last run Nov 16, 2026.
20
runs
15
models
17
configs
3
configs run 2+ times
Hardware
- GPU
- NVIDIA GeForce RTX 5090 32 GB
- CPU
- AMD Ryzen 9 7900X 12-Core Processor 12 cores
- RAM
- 62 GB
- OS
- Fedora Linux 45 (KDE Plasma Desktop Edition)
- Tools
- vLLM 14 · llama.cpp 3 · Strata 3
- Client
- 0.4.99
Models on this machine
One dot per model, at its best config. Up is better answers, right is faster. Time to finish counts every token the model wrote, thinking included, so a model that talks a lot ends up further left.
Best trade-off: quality 60+, and nothing else here is both faster and betterOther models · dot size = number of runs
Medians of each model's best config, 13 models.
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
1 run · 1 config
Qwen3.8-27B-QUASAR-NVFP4
1 run · 1 config
unsloth/Qwen3.8-Flash-Next-GGUF
2 runs · 1 config
★ bestvLLM64k ctxthinking default
×2 runs
87.787.2–88.1
steady
39.1 tok/s
34.9–43.4
AtomicChat/Qwen3.8-Flash-Next-GGUF
1 run · 1 config
esatapedico/Qwen3.8-27B-NVFP4-MTP-GGUF
1 run · 1 config
swift-1.5-iq3_xxs
1 run · 1 config
Qwen3.8-27B QUASAR NVFP4
1 run · 1 config
SC117/Ornith-1.5-35B-A3B-Heretic-MTP-APEX-GGUF
1 run · 1 config
qwen3.6-35b-a3b
1 run · 1 config
qwen3.8-flash-next-iq3_s
2 runs · 2 configs
Compare with another machine
Run history
Nov 16, 2026
04:03Qwen3.8-27B QUASAR NVFP4llama.cpp247 tok/s85.103:34Qwen3.8-Flash-Next UD-IQ1_S×2llama.cpp344 tok/s75.5Oct 2, 2026
17:27swift-1.5-iq3_xxsStrata117 tok/s85.916:16qwen3.8-flash-next-iq3_sStrata112 tok/s79.9Sep 20, 2026
12:27Qwen3.8-Flash-Next-GSQ-RCO-GGUFvLLM27.4 tok/s73.211:40Qwen3.8-Flash-Next-GSQ-RCO-GGUFvLLM36.4 tok/s39.2Show all 20 runs ▾Show fewer ▴
Aug 30, 2026
19:03Qwen3.8-Flash-Next-GGUFvLLM39.7 tok/s86.917:38Qwen3.8-Flash-Next-GGUFvLLM34.9 tok/s88.1Aug 27, 2026
14:42Qwen3.8-Flash-Next-GGUFvLLM43.4 tok/s87.2Aug 26, 2026
04:19Qwen3.8-27B-QUASAR-NVFP4vLLM79.9 tok/s88.3