TEST #3149 RESULTS
03/21/2026 - 11:38 PM
ACCELERATOR
MODEL
148
tokens/s
generation
456
ms
time to first token
2494
tokens/s
prompt
931
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
25.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.3.0: Wed Jan 28 20:55:08 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3159
tokens/s
172
tokens/s
330
ms
pp4096+tg256
2893
tokens/s
101
tokens/s
1.43
sec
pp2048+tg256
3228
tokens/s
136
tokens/s
642
ms
pp2048+tg768
3200
tokens/s
130
tokens/s
647
ms
pp1024+tg1024
3164
tokens/s
150
tokens/s
330
ms
pp1280+tg3072
3002
tokens/s
120
tokens/s
433
ms
pp384+tg1152
2623
tokens/s
167
tokens/s
152
ms
pp64+tg1024
937
tokens/s
181
tokens/s
73
ms
pp16+tg1536
242
tokens/s
173
tokens/s
71
ms