TEST #2162 RESULTS
01/06/2026 - 2:26 PM
ACCELERATOR
MODEL
100
tokens/s
generation
806
ms
time to first token
1473
tokens/s
prompt
568
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Pro 8P+4E
RAM
32GB
OS
Darwin
Kernel Release
25.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.1.0: Mon Oct 20 19:33:00 PDT 2025; root:xnu-12377.41.6~2/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1856
tokens/s
114
tokens/s
561
ms
pp4096+tg256
1561
tokens/s
67.6
tokens/s
2.64
sec
pp2048+tg256
1751
tokens/s
91.1
tokens/s
1.18
sec
pp2048+tg768
1777
tokens/s
87.1
tokens/s
1.16
sec
pp1024+tg1024
1789
tokens/s
102
tokens/s
581
ms
pp1280+tg3072
1776
tokens/s
82.0
tokens/s
730
ms
pp384+tg1152
1670
tokens/s
115
tokens/s
238
ms
pp64+tg1024
852
tokens/s
126
tokens/s
82
ms
pp16+tg1536
223
tokens/s
120
tokens/s
79
ms