TEST #102 RESULTS
03/28/2025 - 9:38 PM
ACCELERATOR
MODEL
141
tokens/s
generation
81
ms
time to first token
16736
tokens/s
prompt
3077
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 7600 6-Core Processor (znver4)
RAM
30.6GB
OS
Linux
Kernel Release
6.8.0-45-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #45~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Sep 11 15:25:05 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
22841
tokens/s
152
tokens/s
51
ms
pp4096+tg256
15226
tokens/s
117
tokens/s
278
ms
pp2048+tg256
19671
tokens/s
137
tokens/s
111
ms
pp2048+tg768
19594
tokens/s
132
tokens/s
112
ms
pp1024+tg1024
22950
tokens/s
143
tokens/s
51
ms
pp1280+tg3072
20853
tokens/s
128
tokens/s
68
ms
pp384+tg1152
22330
tokens/s
150
tokens/s
24
ms
pp64+tg1024
5340
tokens/s
156
tokens/s
18
ms
pp16+tg1536
1821
tokens/s
153
tokens/s
14
ms