TEST #3310 RESULTS
04/15/2026 - 3:50 AM
ACCELERATOR
MODEL
103
tokens/s
generation
369
ms
time to first token
3511
tokens/s
prompt
993
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i9-12900K (alderlake)
RAM
91.8GB
OS
Linux
Kernel Release
6.8.0-107-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #107-Ubuntu SMP PREEMPT_DYNAMIC Fri Mar 13 19:51:50 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4159
tokens/s
106
tokens/s
256
ms
pp4096+tg256
3363
tokens/s
92.1
tokens/s
1.23
sec
pp2048+tg256
3823
tokens/s
102
tokens/s
546
ms
pp2048+tg768
3818
tokens/s
100.0
tokens/s
546
ms
pp1024+tg1024
4101
tokens/s
104
tokens/s
259
ms
pp1280+tg3072
4009
tokens/s
98.2
tokens/s
329
ms
pp384+tg1152
4224
tokens/s
106
tokens/s
100
ms
pp64+tg1024
3027
tokens/s
109
tokens/s
30
ms
pp16+tg1536
1071
tokens/s
108
tokens/s
24
ms