TEST #1748 RESULTS
11/07/2025 - 10:39 PM
ACCELERATOR
MODEL
74.8
tokens/s
generation
376
ms
time to first token
3495
tokens/s
prompt
886
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-14700KF (alderlake)
RAM
31.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4263
tokens/s
75.6
tokens/s
254
ms
pp4096+tg256
3232
tokens/s
70.0
tokens/s
1.28
sec
pp2048+tg256
3805
tokens/s
75.9
tokens/s
552
ms
pp2048+tg768
3808
tokens/s
75.1
tokens/s
551
ms
pp1024+tg1024
4242
tokens/s
75.4
tokens/s
255
ms
pp1280+tg3072
4104
tokens/s
73.1
tokens/s
325
ms
pp384+tg1152
4632
tokens/s
76.3
tokens/s
96
ms
pp64+tg1024
2508
tokens/s
76.1
tokens/s
39
ms
pp16+tg1536
858
tokens/s
76.0
tokens/s
32
ms