Machine
RX 7900 XTX · Core Ultra 9 285K #a6ed9dcd
16 runs of 10 models since Aug 15, 2026. Last run Nov 1, 2026.
16
runs
10
models
15
configs
1
configs run 2+ times
Hardware
- GPU
- AMD Radeon RX 7900 XTX 24 GB
- CPU
- Intel(R) Core(TM) Ultra 9 285K 24 cores
- RAM
- 63 GB
- OS
- Windows 11 IoT Enterprise 25H2
- Tools
- Ollama 16
- Client
- 0.4.98
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, 10 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-mtp-q4_K_M
3 runs · 3 configs
qwen3.8:27b-q4_K_M
2 runs · 1 config
★ bestQ4_K_MOllama64k ctxthinking default
×2 runs
85.885.1–86.5
steady
34.5 tok/s
33.2–35.8
hf.co/ornith-ai/Ornith-1.0-9B-GGUF
1 run · 1 config
Qwen3.8-9B-Q8_0
2 runs · 2 configs
gemma4:26b-a4b-it-qat
3 runs · 3 configs
gemma4:12b
1 run · 1 config
Qwen3.8-9B-Q4_K_M
1 run · 1 config
lfm3:256k
1 run · 1 config
hf.co/LiquidAI/LFM2.5-2.6B-GGUF
1 run · 1 config
hf.co/webAI-Official/TwIL-LM3
1 run · 1 config
Compare with another machine
Run history
Nov 1, 2026
09:45lfm3:256kOllama238 tok/s56.0Aug 24, 2026
06:46Qwen3.8-9B-Q8_0:latestOllama73.5 tok/s76.706:39gemma4:26b-a4b-it-qatOllama106 tok/s73.4Aug 19, 2026
07:18Qwen3.8-9B-Q8_0:latestOllama76.5 tok/s68.9Aug 16, 2026
13:04gemma4:12bOllama63.1 tok/s72.411:18gemma4:26b-a4b-it-qatOllama121 tok/s62.311:14gemma4:26b-a4b-it-qatOllama120 tok/s71.811:02qwen3.8:27b-mtp-q4_K_MOllama42.9 tok/s82.910:54qwen3.8:27b-mtp-q4_K_MOllama41.7 tok/s87.509:29qwen3.8:27b-q4_K_MOllama35.8 tok/s86.509:08qwen3.8:27b-mtp-q4_K_MOllama37.7 tok/s80.2