TEST #3092 RESULTS
03/15/2026 - 2:38 PM
ACCELERATOR
MODEL
53.0
tokens/s
generation
864
ms
time to first token
1506
tokens/s
prompt
452
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i7-11700 @ 2.50GHz (rocketlake)
RAM
2.9GB
OS
Linux
Kernel Release
6.12.73+deb13-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.12.73-1 (2026-02-17)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1501
tokens/s
54.6
tokens/s
700
ms
pp4096+tg256
1416
tokens/s
46.9
tokens/s
2.91
sec
pp2048+tg256
1659
tokens/s
52.1
tokens/s
1.25
sec
pp2048+tg768
1674
tokens/s
51.3
tokens/s
1.24
sec
pp1024+tg1024
1824
tokens/s
53.8
tokens/s
580
ms
pp1280+tg3072
1762
tokens/s
50.2
tokens/s
745
ms
pp384+tg1152
1910
tokens/s
55.4
tokens/s
219
ms
pp64+tg1024
1453
tokens/s
56.5
tokens/s
61
ms
pp16+tg1536
353
tokens/s
56.0
tokens/s
62
ms