TEST #24 RESULTS
03/27/2025 - 10:53 PM
ACCELERATOR
MODEL
216
tokens/s
generation
164
ms
time to first token
9463
tokens/s
prompt
2318
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
125.6GB
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
11613
tokens/s
225
tokens/s
93
ms
pp4096+tg256
6276
tokens/s
193
tokens/s
658
ms
pp2048+tg256
9061
tokens/s
212
tokens/s
231
ms
pp2048+tg768
9058
tokens/s
209
tokens/s
231
ms
pp1024+tg1024
11595
tokens/s
219
tokens/s
93
ms
pp1280+tg3072
10921
tokens/s
206
tokens/s
122
ms
pp384+tg1152
15624
tokens/s
226
tokens/s
29
ms
pp64+tg1024
8425
tokens/s
230
tokens/s
12
ms
pp16+tg1536
2592
tokens/s
227
tokens/s
10
ms