TEST #1943 RESULTS
12/15/2025 - 4:37 PM
ACCELERATOR
MODEL
105
tokens/s
generation
708
ms
time to first token
1627
tokens/s
prompt
622
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Max 8P+2E
RAM
32GB
OS
Darwin
Kernel Release
23.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 23.3.0: Wed Dec 20 21:30:44 PST 2023; root:xnu-10002.81.5~7/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2037
tokens/s
113
tokens/s
511
ms
pp4096+tg256
1845
tokens/s
71.2
tokens/s
2.23
sec
pp2048+tg256
2021
tokens/s
96.3
tokens/s
1.02
sec
pp2048+tg768
2028
tokens/s
92.1
tokens/s
1.02
sec
pp1024+tg1024
1979
tokens/s
108
tokens/s
526
ms
pp1280+tg3072
1951
tokens/s
85.9
tokens/s
665
ms
pp384+tg1152
1735
tokens/s
120
tokens/s
229
ms
pp64+tg1024
846
tokens/s
131
tokens/s
82
ms
pp16+tg1536
203
tokens/s
125
tokens/s
86
ms