TEST #1504 RESULTS
09/11/2025 - 9:46 PM
ACCELERATOR
MODEL
49.9
tokens/s
generation
332
ms
time to first token
3833
tokens/s
prompt
832
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 9800X3D 8-Core Processor
RAM
62.4GB
OS
Linux
Kernel Release
6.12.44-1-lts
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Thu, 28 Aug 2025 15:07:21 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4501
tokens/s
54.1
tokens/s
245
ms
pp4096+tg256
3865
tokens/s
40.4
tokens/s
1.08
sec
pp2048+tg256
4515
tokens/s
48.5
tokens/s
474
ms
pp2048+tg768
4526
tokens/s
47.3
tokens/s
473
ms
pp1024+tg1024
4831
tokens/s
50.7
tokens/s
230
ms
pp1280+tg3072
4601
tokens/s
45.4
tokens/s
297
ms
pp384+tg1152
4910
tokens/s
53.7
tokens/s
97
ms
pp64+tg1024
2143
tokens/s
54.1
tokens/s
46
ms
pp16+tg1536
607
tokens/s
54.6
tokens/s
44
ms