TEST #4 RESULTS
03/19/2025 - 10:06 PM
ACCELERATOR
MODEL
80.4
tokens/s
generation
1.00
sec
time to first token
1204
tokens/s
prompt
459
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Pro 8P+2E
RAM
32GB
OS
Darwin
Kernel Release
24.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.3.0: Thu Jan 2 20:24:16 PST 2025; root:xnu-11215.81.4~3/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.0
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1468
tokens/s
90.4
tokens/s
709
ms
pp4096+tg256
1251
tokens/s
53.6
tokens/s
3.29
sec
pp2048+tg256
1407
tokens/s
73.0
tokens/s
1.47
sec
pp2048+tg768
1404
tokens/s
69.8
tokens/s
1.47
sec
pp1024+tg1024
1448
tokens/s
82.1
tokens/s
718
ms
pp1280+tg3072
1431
tokens/s
65.4
tokens/s
906
ms
pp384+tg1152
1390
tokens/s
92.6
tokens/s
286
ms
pp64+tg1024
784
tokens/s
101
tokens/s
91
ms
pp16+tg1536
250
tokens/s
96.2
tokens/s
73
ms