TEST #3453 RESULTS
05/06/2026 - 7:36 AM
ACCELERATOR
MODEL
42.9
tokens/s
generation
3.94
sec
time to first token
369
tokens/s
prompt
159
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
397
tokens/s
48.3
tokens/s
2.61
sec
pp4096+tg256
284
tokens/s
37.2
tokens/s
14.46
sec
pp2048+tg256
361
tokens/s
41.8
tokens/s
5.70
sec
pp2048+tg768
357
tokens/s
40.7
tokens/s
5.76
sec
pp1024+tg1024
400
tokens/s
43.2
tokens/s
2.58
sec
pp1280+tg3072
399
tokens/s
39.8
tokens/s
3.23
sec
pp384+tg1152
432
tokens/s
44.7
tokens/s
910
ms
pp64+tg1024
413
tokens/s
45.1
tokens/s
176
ms
pp16+tg1536
279
tokens/s
45.3
tokens/s
80
ms