TEST #436 RESULTS
04/06/2025 - 9:55 PM
ACCELERATOR
MODEL
64.7
tokens/s
generation
1.40
sec
time to first token
1468
tokens/s
prompt
407
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 3700X 8-Core Processor (znver2)
RAM
31.3GB
OS
Linux
Kernel Release
6.8.0-55-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #57-Ubuntu SMP PREEMPT_DYNAMIC Wed Feb 12 23:42:21 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
339
tokens/s
64.6
tokens/s
3.03
sec
pp4096+tg256
771
tokens/s
57.3
tokens/s
5.33
sec
pp2048+tg256
1178
tokens/s
62.6
tokens/s
1.75
sec
pp2048+tg768
1784
tokens/s
64.3
tokens/s
1.16
sec
pp1024+tg1024
2038
tokens/s
65.9
tokens/s
517
ms
pp1280+tg3072
2338
tokens/s
63.1
tokens/s
562
ms
pp384+tg1152
2146
tokens/s
65.4
tokens/s
195
ms
pp64+tg1024
2082
tokens/s
69.8
tokens/s
45
ms
pp16+tg1536
535
tokens/s
69.0
tokens/s
44
ms