TEST #3261 RESULTS
04/09/2026 - 1:58 PM
ACCELERATOR
MODEL
38.7
tokens/s
generation
2.24
sec
time to first token
579
tokens/s
prompt
215
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 4P+4E
RAM
16GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Fri Feb 27 19:34:43 PST 2026; root:xnu-11417.140.69.709.8~1/RELEASE_ARM64_T8103
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
718
tokens/s
46.4
tokens/s
1.45
sec
pp4096+tg256
508
tokens/s
27.6
tokens/s
8.10
sec
pp2048+tg256
648
tokens/s
37.3
tokens/s
3.19
sec
pp2048+tg768
652
tokens/s
35.8
tokens/s
3.17
sec
pp1024+tg1024
710
tokens/s
41.7
tokens/s
1.46
sec
pp1280+tg3072
675
tokens/s
32.6
tokens/s
1.92
sec
pp384+tg1152
663
tokens/s
41.6
tokens/s
599
ms
pp64+tg1024
461
tokens/s
43.0
tokens/s
156
ms
pp16+tg1536
171
tokens/s
42.1
tokens/s
112
ms