TEST #1346 RESULTS
08/09/2025 - 12:59 AM
ACCELERATOR
MODEL
23.4
tokens/s
generation
2.20
sec
time to first token
857
tokens/s
prompt
209
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-9500 CPU @ 3.00GHz (skylake)
RAM
79.9GB
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
923
tokens/s
27.1
tokens/s
1.15
sec
pp4096+tg256
442
tokens/s
14.9
tokens/s
9.34
sec
pp2048+tg256
668
tokens/s
19.4
tokens/s
3.12
sec
pp2048+tg768
678
tokens/s
18.4
tokens/s
3.07
sec
pp1024+tg1024
924
tokens/s
23.6
tokens/s
1.14
sec
pp1280+tg3072
863
tokens/s
17.4
tokens/s
1.52
sec
pp384+tg1152
1340
tokens/s
27.7
tokens/s
315
ms
pp64+tg1024
1339
tokens/s
32.5
tokens/s
73
ms
pp16+tg1536
534
tokens/s
29.7
tokens/s
55
ms