ModelDock
3,571.4
132.5
32 GB
ModelDock / model benchmark
Qwen3.8 27B local hardware benchmarks
Compare Qwen3.8 27B reading / prefill and single-stream writing / decode performance across source-backed hardware and clearly labelled projections. Model profile: 27B.
ModelQwen3.8 27B27B · Dataset as of 2026-08-28
Hardware × model dataset
Published Qwen3.8 27B hardware performance
3,250
113.3
32 GB
6,288
112.5
48 GB
7,756
109.3
96 GB
05
Mac Studio M5 Ultra 256 GBEstimate
~1,075
~91
256 GB
06
Mac Studio M5 Ultra 96 GBEstimate
~1,075
~91
96 GB
1,458.3
69.8
24 GB
08
2,692.3
68.8
24 GB
723
65.9
32 GB
745
59.2
128 GB
11
Mac Studio M5 Max 128 GBEstimate
~745
~59.2
128 GB
12
Mac Studio M5 Max 64 GBEstimate
~745
~59.2
64 GB
13
705
55.9
24 GB
14
1,555.6
50.8
16 GB
424.8
49.3
32 GB
462.5
44.3
512 GB
1,225.5
44
128 GB
877.8
43.9
24 GB
19
1,750
43.6
16 GB
20
Mac mini M5 Pro 64 GBEstimate
~373
~29.6
64 GB
21
925.9
28.3
16 GB
22
779.5
27.4
16 GB
23
~206
~16.4
32 GB
Reading / prefill
Prefill · tok/s
Writing / decode
Decode · single stream · tok/s
Evidence
Rows use the same hardware × model projection as the live Benchmark and Ranking interfaces. Estimates remain visibly marked.