TEST #3190 RESULTS
03/25/2026 - 6:53 AM
ACCELERATOR
MODEL
22.0
tokens/s
generation
5.64
sec
time to first token
224
tokens/s
prompt
96
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Pro 5P+6E
RAM
18GB
OS
Darwin
Kernel Release
23.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 23.4.0: Fri Mar 15 00:12:25 PDT 2024; root:xnu-10063.101.17~1/RELEASE_ARM64_T6030
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
253
tokens/s
22.8
tokens/s
4.09
sec
pp4096+tg256
224
tokens/s
18.2
tokens/s
18.32
sec
pp2048+tg256
244
tokens/s
21.3
tokens/s
8.45
sec
pp2048+tg768
244
tokens/s
20.9
tokens/s
8.46
sec
pp1024+tg1024
251
tokens/s
22.6
tokens/s
4.12
sec
pp1280+tg3072
250
tokens/s
20.3
tokens/s
5.18
sec
pp384+tg1152
252
tokens/s
23.7
tokens/s
1.56
sec
pp64+tg1024
205
tokens/s
24.4
tokens/s
351
ms
pp16+tg1536
89
tokens/s
24.1
tokens/s
218
ms