TEST #1292 RESULTS
08/03/2025 - 8:39 AM
ACCELERATOR
MODEL
153
tokens/s
generation
442
ms
time to first token
2551
tokens/s
prompt
959
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Max 8P+4E
RAM
32GB
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 19:54:25 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3243
tokens/s
175
tokens/s
322
ms
pp4096+tg256
3008
tokens/s
103
tokens/s
1.37
sec
pp2048+tg256
3274
tokens/s
140
tokens/s
633
ms
pp2048+tg768
3285
tokens/s
134
tokens/s
630
ms
pp1024+tg1024
3257
tokens/s
158
tokens/s
320
ms
pp1280+tg3072
3246
tokens/s
125
tokens/s
400
ms
pp384+tg1152
2553
tokens/s
173
tokens/s
156
ms
pp64+tg1024
820
tokens/s
189
tokens/s
83
ms
pp16+tg1536
277
tokens/s
180
tokens/s
63
ms