TEST #3991 RESULTS
08/03/2026 - 7:30 PM
ACCELERATOR
MODEL
29.3
tokens/s
generation
4.21
sec
time to first token
359
tokens/s
prompt
136
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 3950X 16-Core Processor (znver2)
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
385
tokens/s
30.9
tokens/s
2.69
sec
pp4096+tg256
273
tokens/s
26.2
tokens/s
15.04
sec
pp2048+tg256
339
tokens/s
29.1
tokens/s
6.07
sec
pp2048+tg768
337
tokens/s
26.6
tokens/s
6.12
sec
pp1024+tg1024
302
tokens/s
29.6
tokens/s
3.43
sec
pp1280+tg3072
383
tokens/s
27.9
tokens/s
3.38
sec
pp384+tg1152
418
tokens/s
30.8
tokens/s
951
ms
pp64+tg1024
427
tokens/s
31.5
tokens/s
181
ms
pp16+tg1536
367
tokens/s
31.0
tokens/s
74
ms