TEST #736 RESULTS
04/18/2025 - 2:22 PM
ACCELERATOR
MODEL
47.9
tokens/s
generation
967
ms
time to first token
1770
tokens/s
prompt
444
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 7535HS with Radeon Graphics (znver3)
RAM
14.9GB
OS
Linux
Kernel Release
6.14.2-arch1-1
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Thu, 10 Apr 2025 18:43:59 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3455
tokens/s
102
tokens/s
306
ms
pp4096+tg256
1978
tokens/s
81.5
tokens/s
2.08
sec
pp2048+tg256
2774
tokens/s
93.9
tokens/s
749
ms
pp2048+tg768
2774
tokens/s
92.2
tokens/s
749
ms
pp1024+tg1024
3444
tokens/s
14.2
tokens/s
307
ms
pp1280+tg3072
408
tokens/s
11.1
tokens/s
3.22
sec
pp384+tg1152
472
tokens/s
12.1
tokens/s
895
ms
pp64+tg1024
466
tokens/s
12.3
tokens/s
217
ms
pp16+tg1536
156
tokens/s
12.2
tokens/s
181
ms