TEST #945 RESULTS
05/25/2025 - 9:54 AM
ACCELERATOR
MODEL
111
tokens/s
generation
684
ms
time to first token
1736
tokens/s
prompt
655
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 8P+4E
RAM
24GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:53:27 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2200
tokens/s
125
tokens/s
473
ms
pp4096+tg256
1868
tokens/s
74.2
tokens/s
2.21
sec
pp2048+tg256
2049
tokens/s
100
tokens/s
1.01
sec
pp2048+tg768
2041
tokens/s
95.9
tokens/s
1.01
sec
pp1024+tg1024
2102
tokens/s
113
tokens/s
495
ms
pp1280+tg3072
2111
tokens/s
90.3
tokens/s
615
ms
pp384+tg1152
1893
tokens/s
127
tokens/s
210
ms
pp64+tg1024
1087
tokens/s
138
tokens/s
65
ms
pp16+tg1536
271
tokens/s
132
tokens/s
65
ms