TEST #273 RESULTS
04/05/2025 - 8:24 AM
ACCELERATOR
MODEL
170
tokens/s
generation
339
ms
time to first token
3272
tokens/s
prompt
1179
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Ultra 16P+8E
RAM
192GB
OS
Darwin
Kernel Release
24.2.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.2.0: Sun Nov 3 20:57:50 PST 2024; root:xnu-11215.60.405~54/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4178
tokens/s
187
tokens/s
250
ms
pp4096+tg256
4206
tokens/s
123
tokens/s
982
ms
pp2048+tg256
4414
tokens/s
159
tokens/s
470
ms
pp2048+tg768
4491
tokens/s
154
tokens/s
463
ms
pp1024+tg1024
4018
tokens/s
175
tokens/s
261
ms
pp1280+tg3072
4149
tokens/s
145
tokens/s
314
ms
pp384+tg1152
3058
tokens/s
188
tokens/s
130
ms
pp64+tg1024
739
tokens/s
201
tokens/s
93
ms
pp16+tg1536
197
tokens/s
195
tokens/s
86
ms