TEST #1240 RESULTS
07/23/2025 - 9:14 AM
ACCELERATOR
MODEL
61.5
tokens/s
generation
526
ms
time to first token
2535
tokens/s
prompt
667
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
2998
tokens/s
63.3
tokens/s
357
ms
pp4096+tg256
2292
tokens/s
55.8
tokens/s
1.80
sec
pp2048+tg256
2716
tokens/s
60.8
tokens/s
771
ms
pp2048+tg768
2713
tokens/s
60.1
tokens/s
771
ms
pp1024+tg1024
2993
tokens/s
62.3
tokens/s
358
ms
pp1280+tg3072
2890
tokens/s
59.0
tokens/s
459
ms
pp384+tg1152
3295
tokens/s
63.6
tokens/s
132
ms
pp64+tg1024
2122
tokens/s
64.6
tokens/s
45
ms
pp16+tg1536
792
tokens/s
63.8
tokens/s
35
ms