TEST #120 RESULTS
03/28/2025 - 10:00 PM
ACCELERATOR
MODEL
321
tokens/s
generation
95
ms
time to first token
14233
tokens/s
prompt
3638
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2690 v4 @ 2.60GHz (broadwell)
RAM
62.7GB
OS
Linux
Kernel Release
5.15.0-134-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #145~20.04.1-Ubuntu SMP Mon Feb 17 13:27:16 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
17299
tokens/s
313
tokens/s
62
ms
pp4096+tg256
12595
tokens/s
276
tokens/s
329
ms
pp2048+tg256
16258
tokens/s
315
tokens/s
129
ms
pp2048+tg768
16288
tokens/s
310
tokens/s
129
ms
pp1024+tg1024
18771
tokens/s
327
tokens/s
58
ms
pp1280+tg3072
17422
tokens/s
304
tokens/s
77
ms
pp384+tg1152
19523
tokens/s
344
tokens/s
22
ms
pp64+tg1024
9497
tokens/s
357
tokens/s
9
ms
pp16+tg1536
446
tokens/s
344
tokens/s
39
ms