TEST #1208 RESULTS
07/11/2025 - 5:42 PM
ACCELERATOR
MODEL
6.1
tokens/s
generation
73.94
sec
time to first token
18
tokens/s
prompt
11
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i7-1185G7 @ 3.00GHz (tigerlake)
RAM
15.3GB
OS
Linux
Kernel Release
5.19.0-38-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #39-Ubuntu SMP PREEMPT_DYNAMIC Fri Mar 17 17:33:16 UTC 2023
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
20
tokens/s
6.5
tokens/s
50.72
sec
pp4096+tg256
18
tokens/s
5.3
tokens/s
233.49
sec
pp2048+tg256
18
tokens/s
6.0
tokens/s
113.67
sec
pp2048+tg768
18
tokens/s
5.9
tokens/s
113.21
sec
pp1024+tg1024
18
tokens/s
6.3
tokens/s
56.41
sec
pp1280+tg3072
18
tokens/s
5.7
tokens/s
71.77
sec
pp384+tg1152
18
tokens/s
6.5
tokens/s
21.42
sec
pp64+tg1024
18
tokens/s
6.6
tokens/s
3.68
sec
pp16+tg1536
17
tokens/s
6.5
tokens/s
1.06
sec