TEST #614 RESULTS
04/08/2025 - 8:34 PM
ACCELERATOR
MODEL
34.6
tokens/s
generation
216
ms
time to first token
6502
tokens/s
prompt
1013
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
31.1GB
OS
Linux
Kernel Release
6.14.0-1-default
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Tue Mar 25 06:23:18 UTC 2025 (ab99d4c)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
9498
tokens/s
53.7
tokens/s
127
ms
pp4096+tg256
5679
tokens/s
14.2
tokens/s
790
ms
pp2048+tg256
7974
tokens/s
24.8
tokens/s
285
ms
pp2048+tg768
7899
tokens/s
23.0
tokens/s
288
ms
pp1024+tg1024
8762
tokens/s
29.1
tokens/s
136
ms
pp1280+tg3072
8210
tokens/s
20.9
tokens/s
177
ms
pp384+tg1152
7461
tokens/s
42.8
tokens/s
66
ms
pp64+tg1024
2371
tokens/s
55.1
tokens/s
39
ms
pp16+tg1536
661
tokens/s
47.7
tokens/s
37
ms