TEST #1261 RESULTS
07/27/2025 - 6:20 PM
ACCELERATOR
MODEL
183
tokens/s
generation
301
ms
time to first token
3878
tokens/s
prompt
1330
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
64GB
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
4931
tokens/s
203
tokens/s
213
ms
pp4096+tg256
4246
tokens/s
124
tokens/s
973
ms
pp2048+tg256
4798
tokens/s
166
tokens/s
433
ms
pp2048+tg768
4746
tokens/s
159
tokens/s
437
ms
pp1024+tg1024
4908
tokens/s
185
tokens/s
213
ms
pp1280+tg3072
4802
tokens/s
150
tokens/s
272
ms
pp384+tg1152
4162
tokens/s
212
tokens/s
96
ms
pp64+tg1024
1808
tokens/s
230
tokens/s
39
ms
pp16+tg1536
505
tokens/s
217
tokens/s
36
ms