TEST #2222 RESULTS
01/10/2026 - 12:25 PM
ACCELERATOR
MODEL
59.1
tokens/s
generation
822
ms
time to first token
1540
tokens/s
prompt
480
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon W-2133 CPU @ 3.60GHz (skylake-avx512)
RAM
30.9GB
OS
Linux
Kernel Release
6.14.0-37-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #37~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Nov 20 10:25:38 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1844
tokens/s
62.9
tokens/s
571
ms
pp4096+tg256
1521
tokens/s
51.2
tokens/s
2.71
sec
pp2048+tg256
1717
tokens/s
57.9
tokens/s
1.21
sec
pp2048+tg768
1712
tokens/s
56.8
tokens/s
1.22
sec
pp1024+tg1024
1811
tokens/s
59.8
tokens/s
581
ms
pp1280+tg3072
1744
tokens/s
54.2
tokens/s
751
ms
pp384+tg1152
1649
tokens/s
62.1
tokens/s
248
ms
pp64+tg1024
1378
tokens/s
64.4
tokens/s
61
ms
pp16+tg1536
484
tokens/s
63.0
tokens/s
48
ms