TEST #1239 RESULTS
07/23/2025 - 9:11 AM
ACCELERATOR
MODEL
60.0
tokens/s
generation
654
ms
time to first token
1989
tokens/s
prompt
567
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2699 v3 @ 2.30GHz (haswell)
RAM
62.8GB
OS
Linux
Kernel Release
6.15.7-arch1-1
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Thu, 17 Jul 2025 21:05:29 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2371
tokens/s
63.0
tokens/s
448
ms
pp4096+tg256
1884
tokens/s
55.3
tokens/s
2.19
sec
pp2048+tg256
2187
tokens/s
59.8
tokens/s
953
ms
pp2048+tg768
2167
tokens/s
58.9
tokens/s
962
ms
pp1024+tg1024
2335
tokens/s
60.8
tokens/s
455
ms
pp1280+tg3072
2233
tokens/s
57.0
tokens/s
590
ms
pp384+tg1152
2469
tokens/s
61.4
tokens/s
172
ms
pp64+tg1024
1965
tokens/s
62.3
tokens/s
48
ms
pp16+tg1536
287
tokens/s
61.7
tokens/s
71
ms