TEST #17 RESULTS
03/24/2025 - 9:01 PM
ACCELERATOR
MODEL
40.7
tokens/s
generation
2.11
sec
time to first token
615
tokens/s
prompt
228
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 4P+4E
RAM
8GB
OS
Darwin
Kernel Release
23.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 23.3.0: Wed Dec 20 21:30:27 PST 2023; root:xnu-10002.81.5~7/RELEASE_ARM64_T8103
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
726
tokens/s
46.3
tokens/s
1.43
sec
pp4096+tg256
552
tokens/s
27.0
tokens/s
7.46
sec
pp2048+tg256
676
tokens/s
37.1
tokens/s
3.06
sec
pp2048+tg768
678
tokens/s
35.3
tokens/s
3.05
sec
pp1024+tg1024
727
tokens/s
41.7
tokens/s
1.43
sec
pp1280+tg3072
719
tokens/s
33.1
tokens/s
1.80
sec
pp384+tg1152
729
tokens/s
47.0
tokens/s
545
ms
pp64+tg1024
532
tokens/s
50.7
tokens/s
137
ms
pp16+tg1536
196
tokens/s
48.3
tokens/s
98
ms