TEST #3393 RESULTS
04/26/2026 - 8:12 PM
ACCELERATOR
MODEL
76.6
tokens/s
generation
1.19
sec
time to first token
1101
tokens/s
prompt
413
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 4P+6E
RAM
24GB
OS
Darwin
Kernel Release
25.2.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.2.0: Tue Nov 18 21:08:48 PST 2025; root:xnu-12377.61.12~1/RELEASE_ARM64_T8132
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1320
tokens/s
84.6
tokens/s
787
ms
pp4096+tg256
967
tokens/s
54.8
tokens/s
4.26
sec
pp2048+tg256
1193
tokens/s
71.5
tokens/s
1.73
sec
pp2048+tg768
1205
tokens/s
68.8
tokens/s
1.71
sec
pp1024+tg1024
1286
tokens/s
77.6
tokens/s
808
ms
pp1280+tg3072
1290
tokens/s
64.9
tokens/s
1.00
sec
pp384+tg1152
1328
tokens/s
86.0
tokens/s
300
ms
pp64+tg1024
1041
tokens/s
92.0
tokens/s
71
ms
pp16+tg1536
281
tokens/s
88.7
tokens/s
67
ms