TEST #86 RESULTS
03/28/2025 - 9:21 PM
ACCELERATOR
MODEL
131
tokens/s
generation
68
ms
time to first token
19620
tokens/s
prompt
3350
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 9354 32-Core Processor (znver4)
RAM
755.5GB
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
26923
tokens/s
137
tokens/s
45
ms
pp4096+tg256
19382
tokens/s
121
tokens/s
220
ms
pp2048+tg256
24363
tokens/s
132
tokens/s
92
ms
pp2048+tg768
24423
tokens/s
129
tokens/s
92
ms
pp1024+tg1024
27911
tokens/s
131
tokens/s
44
ms
pp1280+tg3072
22793
tokens/s
126
tokens/s
64
ms
pp384+tg1152
23513
tokens/s
133
tokens/s
24
ms
pp64+tg1024
5648
tokens/s
135
tokens/s
18
ms
pp16+tg1536
1622
tokens/s
134
tokens/s
16
ms