TEST #634 RESULTS
04/09/2025 - 8:58 AM
ACCELERATOR
MODEL
57.4
tokens/s
generation
540
ms
time to first token
2457
tokens/s
prompt
639
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i9-13980HX (alderlake)
RAM
191.7GB
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
2948
tokens/s
59.0
tokens/s
365
ms
pp4096+tg256
2227
tokens/s
54.4
tokens/s
1.86
sec
pp2048+tg256
2674
tokens/s
57.0
tokens/s
782
ms
pp2048+tg768
2658
tokens/s
57.1
tokens/s
788
ms
pp1024+tg1024
2979
tokens/s
57.3
tokens/s
361
ms
pp1280+tg3072
2845
tokens/s
56.6
tokens/s
469
ms
pp384+tg1152
3242
tokens/s
57.5
tokens/s
135
ms
pp64+tg1024
2054
tokens/s
58.4
tokens/s
50
ms
pp16+tg1536
487
tokens/s
59.5
tokens/s
50
ms