TEST #3635 RESULTS
05/27/2026 - 7:00 PM
ACCELERATOR
MODEL
39.6
tokens/s
generation
825
ms
time to first token
1624
tokens/s
prompt
427
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i5-11400F @ 2.60GHz (rocketlake)
RAM
15.7GB
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
1914
tokens/s
40.2
tokens/s
560
ms
pp4096+tg256
1458
tokens/s
36.8
tokens/s
2.84
sec
pp2048+tg256
1732
tokens/s
39.6
tokens/s
1.21
sec
pp2048+tg768
1734
tokens/s
38.5
tokens/s
1.21
sec
pp1024+tg1024
1909
tokens/s
40.1
tokens/s
562
ms
pp1280+tg3072
1860
tokens/s
38.6
tokens/s
715
ms
pp384+tg1152
2141
tokens/s
40.7
tokens/s
204
ms
pp64+tg1024
1413
tokens/s
41.1
tokens/s
69
ms
pp16+tg1536
458
tokens/s
40.8
tokens/s
60
ms