TEST #3517 RESULTS
05/12/2026 - 8:56 AM
ACCELERATOR
MODEL
65.0
tokens/s
generation
1.46
sec
time to first token
907
tokens/s
prompt
343
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 4P+4E
RAM
24GB
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 20:56:34 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T8122
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1095
tokens/s
75.0
tokens/s
950
ms
pp4096+tg256
806
tokens/s
46.3
tokens/s
5.10
sec
pp2048+tg256
935
tokens/s
59.6
tokens/s
2.21
sec
pp2048+tg768
928
tokens/s
56.4
tokens/s
2.22
sec
pp1024+tg1024
1029
tokens/s
64.4
tokens/s
1.01
sec
pp1280+tg3072
1098
tokens/s
55.1
tokens/s
1.18
sec
pp384+tg1152
1139
tokens/s
74.9
tokens/s
349
ms
pp64+tg1024
862
tokens/s
78.6
tokens/s
86
ms
pp16+tg1536
275
tokens/s
74.9
tokens/s
70
ms