Machine

RX 7800 Xt · Ryzen 5 5600 #67119987

18 runs of 8 models since Nov 10, 2026. Last run Nov 19, 2026.

ShareRedditX
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
All AMD Radeon RX 7800 XT benchmarks

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
below 20 tok/s: slower than you read4050607080901002050100tok/s (log scale) →quality →

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.

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
★ bestllama.cpp64k ctxthinking defaulteffort xhigh
85.2
1 rated run
51.4 tok/s
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
★ bestornith-35b-a3b-think8k ctxthinking on
83.2
1 rated run
89.7 tok/s
qwen3.8-27b
2 runs · 2 configs
★ bestyasakani-hip-think8k ctxthinking on
81.3
1 rated run
56.7 tok/s
Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp
2 runs · 2 configs
★ bestqwen-mtp-think8k ctxthinking on
80.1
1 rated run
48.6 tok/s
qwen3.8-flash-next-sync-s
2 runs · 2 configs
★ bestsync-strata8k ctxthinking on
67.3
1 rated run
39.4 tok/s
Qwen3.8-Flash-Next-Avalon-IQ3KS-00001-of-00033
1 run · 1 config
★ bestllama.cpp8k ctxthinking defaulteffort xhigh
66.4
1 rated run
33.9 tok/s

Compare with another machine

Run history