TEST #117 RESULTS
03/28/2025 - 9:55 PM
ACCELERATOR
MODEL
51.3
tokens/s
generation
1.04
sec
time to first token
1223
tokens/s
prompt
392
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2687W 0 @ 3.10GHz (sandybridge)
RAM
503.8GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 15 19:18:46 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1428
tokens/s
55.9
tokens/s
735
ms
pp4096+tg256
1210
tokens/s
44.0
tokens/s
3.41
sec
pp2048+tg256
1341
tokens/s
49.9
tokens/s
1.55
sec
pp2048+tg768
1332
tokens/s
49.2
tokens/s
1.56
sec
pp1024+tg1024
1403
tokens/s
52.4
tokens/s
748
ms
pp1280+tg3072
1363
tokens/s
46.8
tokens/s
957
ms
pp384+tg1152
1375
tokens/s
53.7
tokens/s
296
ms
pp64+tg1024
1145
tokens/s
55.5
tokens/s
72
ms
pp16+tg1536
409
tokens/s
54.5
tokens/s
55
ms