Machine
RX 7800 Xt · Ryzen 5 5600 #67119987
18 runs of 8 models since Nov 10, 2026. Last run Nov 19, 2026.
18
runs
8
models
13
configs
2
configs run 2+ times
Hardware
- GPU
- AMD Radeon RX 7800 XT 16 GB
- CPU
- AMD Ryzen 5 5600 6-Core Processor 6 cores
- RAM
- 31 GB
- OS
- Linux Mint 22.3
- Tools
- llama.cpp 9 · sync-strata 2 · ornith-35b-a3b-think 1 · qwen-mtp-think 1 · yasakani-hip-think 1 · rdna-fast-huihui-think 1 · rdna-fast-huihui 1 · rdna-fast 1 · yasakani-hip-greedy 1
- 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, 8 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
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_S-mtp
4 runs · 3 configs
★ bestllama.cpp64k ctxthinking defaulteffort xhigh
×2 runs
89.188.4–89.8
steady
43.5 tok/s
40.5–46.5
Qwen3.8-27B-Coder390-Q2LynnStyle
1 run · 1 config
Swift-1.5-Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp
5 runs · 1 config
★ bestllama.cpp64k ctxthinking defaulteffort xhigh
×5 runs
85.180.4–89.4
varies1 slow run
46.9 tok/s
29.2–50.5
Ornith-35B-A3B
1 run · 1 config
qwen3.8-27b
2 runs · 2 configs
Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp
2 runs · 2 configs
qwen3.8-flash-next-sync-s
2 runs · 2 configs
Qwen3.8-Flash-Next-Avalon-IQ3KS-00001-of-00033
1 run · 1 config
Compare with another machine
Run history
Nov 19, 2026
15:52qwen3.8-flash-next-sync-ssync-strata39.4 tok/s67.315:39qwen3.8-flash-next-sync-ssync-strata42.7 tok/s51.5Nov 16, 2026
07:46Ornith-35B-A3Bornith-35b-a3b-think89.7 tok/s83.207:37Qwen3.8-27B-GSQ-RCO-IQ3_S-mtpqwen-mtp-think48.6 tok/s80.107:29qwen3.8-27byasakani-hip-think56.7 tok/s81.307:21Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_S-mtprdna-fast-huihui-think51.5 tok/s68.607:10Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_S-mtprdna-fast-huihui53.5 tok/s60.406:37Qwen3.8-27B-GSQ-RCO-IQ3_S-mtprdna-fast53.1 tok/s48.706:27qwen3.8-27byasakani-hip-greedy62.6 tok/s72.2Nov 12, 2026
14:28Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_S-mtp×2llama.cpp43.5 tok/s89.1