TEST #1729 RESULTS
11/04/2025 - 12:30 AM
ACCELERATOR
MODEL
40.9
tokens/s
generation
1.52
sec
time to first token
833
tokens/s
prompt
282
LocalScore
HOW YOU STACK UP
Explore All ResultsLLaMA v2 - Q4_0
SYSTEM
CPU
Intel Xeon Gold 6146 CPU @ 3.20GHz (skylake-avx512)
RAM
250.5GB
OS
Linux
Kernel Release
5.15.0-143-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #153-Ubuntu SMP Fri Jun 13 19:10:45 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
980
tokens/s
42.3
tokens/s
1.07
sec
pp4096+tg256
817
tokens/s
27.6
tokens/s
5.05
sec
pp2048+tg256
911
tokens/s
37.4
tokens/s
2.28
sec
pp2048+tg768
919
tokens/s
36.2
tokens/s
2.25
sec
pp1024+tg1024
976
tokens/s
42.5
tokens/s
1.07
sec
pp1280+tg3072
938
tokens/s
33.6
tokens/s
1.39
sec
pp384+tg1152
932
tokens/s
47.4
tokens/s
430
ms
pp64+tg1024
697
tokens/s
51.6
tokens/s
109
ms
pp16+tg1536
331
tokens/s
49.4
tokens/s
65
ms