TEST #61 RESULTS
03/28/2025 - 8:53 PM
ACCELERATOR
MODEL
308
tokens/s
generation
80
ms
time to first token
15370
tokens/s
prompt
3892
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7713 64-Core Processor (znver3)
RAM
1007.7GB
OS
Linux
Kernel Release
6.5.0-28-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #29~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Apr 4 14:39:20 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
18601
tokens/s
316
tokens/s
58
ms
pp4096+tg256
16096
tokens/s
281
tokens/s
258
ms
pp2048+tg256
18675
tokens/s
305
tokens/s
113
ms
pp2048+tg768
18795
tokens/s
300
tokens/s
112
ms
pp1024+tg1024
19948
tokens/s
310
tokens/s
55
ms
pp1280+tg3072
18900
tokens/s
296
tokens/s
71
ms
pp384+tg1152
18898
tokens/s
320
tokens/s
23
ms
pp64+tg1024
7527
tokens/s
324
tokens/s
12
ms
pp16+tg1536
892
tokens/s
320
tokens/s
21
ms