TEST #3520 RESULTS
05/12/2026 - 9:22 PM
ACCELERATOR
MODEL
165
tokens/s
generation
364
ms
time to first token
3347
tokens/s
prompt
1150
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 10P+4E
RAM
36GB
OS
Darwin
Kernel Release
25.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.4.0: Thu Mar 19 19:33:25 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4153
tokens/s
182
tokens/s
252
ms
pp4096+tg256
3402
tokens/s
116
tokens/s
1.21
sec
pp2048+tg256
3901
tokens/s
152
tokens/s
531
ms
pp2048+tg768
3919
tokens/s
146
tokens/s
529
ms
pp1024+tg1024
4071
tokens/s
168
tokens/s
257
ms
pp1280+tg3072
4031
tokens/s
139
tokens/s
323
ms
pp384+tg1152
3908
tokens/s
189
tokens/s
103
ms
pp64+tg1024
2060
tokens/s
202
tokens/s
36
ms
pp16+tg1536
681
tokens/s
193
tokens/s
28
ms