TEST #3076 RESULTS
03/13/2026 - 8:58 PM
ACCELERATOR
MODEL
83.0
tokens/s
generation
740
ms
time to first token
1537
tokens/s
prompt
556
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
25.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.3.0: Wed Jan 28 20:54:46 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1982
tokens/s
94.8
tokens/s
526
ms
pp4096+tg256
1825
tokens/s
63.9
tokens/s
2.26
sec
pp2048+tg256
1969
tokens/s
68.6
tokens/s
1.05
sec
pp2048+tg768
1957
tokens/s
71.4
tokens/s
1.06
sec
pp1024+tg1024
1914
tokens/s
81.1
tokens/s
547
ms
pp1280+tg3072
1704
tokens/s
69.6
tokens/s
767
ms
pp384+tg1152
1711
tokens/s
84.6
tokens/s
232
ms
pp64+tg1024
590
tokens/s
123
tokens/s
116
ms
pp16+tg1536
179
tokens/s
89.9
tokens/s
97
ms