TEST #1461 RESULTS
08/30/2025 - 10:04 AM
ACCELERATOR
MODEL
62.6
tokens/s
generation
1.07
sec
time to first token
1100
tokens/s
prompt
401
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Ultra 24P+8E
RAM
512GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Jul 14 11:30:55 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6031
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1337
tokens/s
67.8
tokens/s
781
ms
pp4096+tg256
1238
tokens/s
48.8
tokens/s
3.33
sec
pp2048+tg256
1308
tokens/s
59.5
tokens/s
1.58
sec
pp2048+tg768
1310
tokens/s
58.1
tokens/s
1.58
sec
pp1024+tg1024
1323
tokens/s
63.9
tokens/s
789
ms
pp1280+tg3072
1297
tokens/s
55.7
tokens/s
1.00
sec
pp384+tg1152
1235
tokens/s
68.3
tokens/s
325
ms
pp64+tg1024
662
tokens/s
71.6
tokens/s
110
ms
pp16+tg1536
188
tokens/s
70.0
tokens/s
98
ms