TEST #1518 RESULTS
09/15/2025 - 10:54 AM
ACCELERATOR
MODEL
92.1
tokens/s
generation
282
ms
time to first token
4698
tokens/s
prompt
1153
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i9-13900KF (alderlake)
RAM
31.2GB
OS
Linux
Kernel Release
6.15.11-2-MANJARO
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Sat, 30 Aug 2025 05:31:45 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5739
tokens/s
92.2
tokens/s
189
ms
pp4096+tg256
4244
tokens/s
91.0
tokens/s
976
ms
pp2048+tg256
4683
tokens/s
84.1
tokens/s
450
ms
pp2048+tg768
5334
tokens/s
89.3
tokens/s
394
ms
pp1024+tg1024
5770
tokens/s
93.5
tokens/s
188
ms
pp1280+tg3072
6352
tokens/s
88.9
tokens/s
213
ms
pp384+tg1152
6744
tokens/s
93.0
tokens/s
67
ms
pp64+tg1024
2265
tokens/s
98.6
tokens/s
39
ms
pp16+tg1536
1151
tokens/s
98.1
tokens/s
24
ms