TEST #509 RESULTS
04/07/2025 - 6:47 PM
ACCELERATOR
MODEL
69.1
tokens/s
generation
345
ms
time to first token
3985
tokens/s
prompt
927
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i7-13700K (alderlake)
RAM
63.7GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4909
tokens/s
69.9
tokens/s
223
ms
pp4096+tg256
3367
tokens/s
66.1
tokens/s
1.23
sec
pp2048+tg256
4232
tokens/s
68.5
tokens/s
499
ms
pp2048+tg768
4231
tokens/s
68.1
tokens/s
499
ms
pp1024+tg1024
4894
tokens/s
69.2
tokens/s
224
ms
pp1280+tg3072
4769
tokens/s
68.0
tokens/s
283
ms
pp384+tg1152
5534
tokens/s
70.1
tokens/s
83
ms
pp64+tg1024
3043
tokens/s
70.9
tokens/s
35
ms
pp16+tg1536
881
tokens/s
70.7
tokens/s
32
ms