TEST #1453 RESULTS
08/29/2025 - 7:59 PM
ACCELERATOR
MODEL
43.5
tokens/s
generation
1.07
sec
time to first token
1265
tokens/s
prompt
373
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i7-11800H @ 2.30GHz (tigerlake)
RAM
62.6GB
OS
Linux
Kernel Release
6.14.0-29-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #29~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Aug 14 16:52:50 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1421
tokens/s
44.1
tokens/s
741
ms
pp4096+tg256
1154
tokens/s
39.4
tokens/s
3.57
sec
pp2048+tg256
1298
tokens/s
43.5
tokens/s
1.60
sec
pp2048+tg768
1308
tokens/s
43.2
tokens/s
1.59
sec
pp1024+tg1024
1418
tokens/s
44.9
tokens/s
742
ms
pp1280+tg3072
1355
tokens/s
41.9
tokens/s
965
ms
pp384+tg1152
1580
tokens/s
44.7
tokens/s
263
ms
pp64+tg1024
1361
tokens/s
45.4
tokens/s
66
ms
pp16+tg1536
494
tokens/s
45.0
tokens/s
51
ms