TEST #1403 RESULTS
08/20/2025 - 1:16 PM
ACCELERATOR
MODEL
8.0
tokens/s
generation
52.91
sec
time to first token
24
tokens/s
prompt
15
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i5-12600HX (alderlake)
RAM
31GB
OS
Linux
Kernel Release
6.12.33+kali-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Kali 6.12.33-1kali1 (2025-06-25)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
30
tokens/s
10.0
tokens/s
34.54
sec
pp4096+tg256
24
tokens/s
7.8
tokens/s
171.00
sec
pp2048+tg256
27
tokens/s
9.2
tokens/s
76.34
sec
pp2048+tg768
27
tokens/s
8.4
tokens/s
77.07
sec
pp1024+tg1024
24
tokens/s
8.1
tokens/s
42.21
sec
pp1280+tg3072
24
tokens/s
6.4
tokens/s
52.63
sec
pp384+tg1152
21
tokens/s
7.4
tokens/s
18.34
sec
pp64+tg1024
21
tokens/s
7.6
tokens/s
3.10
sec
pp16+tg1536
20
tokens/s
7.5
tokens/s
925
ms