TEST #1390 RESULTS
08/16/2025 - 9:31 AM
ACCELERATOR
MODEL
38.4
tokens/s
generation
1.23
sec
time to first token
1094
tokens/s
prompt
325
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800HS with Radeon Graphics (znver3)
RAM
15GB
OS
Linux
Kernel Release
6.8.0-71-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #71-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 22 16:52:38 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1152
tokens/s
38.0
tokens/s
913
ms
pp4096+tg256
1006
tokens/s
34.3
tokens/s
4.10
sec
pp2048+tg256
1155
tokens/s
38.1
tokens/s
1.80
sec
pp2048+tg768
1117
tokens/s
37.4
tokens/s
1.86
sec
pp1024+tg1024
1255
tokens/s
39.3
tokens/s
840
ms
pp1280+tg3072
1199
tokens/s
36.3
tokens/s
1.09
sec
pp384+tg1152
1385
tokens/s
40.4
tokens/s
300
ms
pp64+tg1024
1222
tokens/s
41.2
tokens/s
74
ms
pp16+tg1536
354
tokens/s
40.6
tokens/s
66
ms