TEST #2205 RESULTS
01/09/2026 - 11:54 AM
ACCELERATOR
MODEL
15.2
tokens/s
generation
8.93
sec
time to first token
144
tokens/s
prompt
63
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 4P+4E
RAM
16GB
OS
Darwin
Kernel Release
25.2.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.2.0: Tue Nov 18 21:09:34 PST 2025; root:xnu-12377.61.12~1/RELEASE_ARM64_T8112
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
160
tokens/s
16.4
tokens/s
6.48
sec
pp4096+tg256
141
tokens/s
12.1
tokens/s
29.13
sec
pp2048+tg256
153
tokens/s
14.5
tokens/s
13.42
sec
pp2048+tg768
153
tokens/s
14.2
tokens/s
13.42
sec
pp1024+tg1024
159
tokens/s
15.5
tokens/s
6.49
sec
pp1280+tg3072
158
tokens/s
13.7
tokens/s
8.16
sec
pp384+tg1152
161
tokens/s
16.5
tokens/s
2.44
sec
pp64+tg1024
147
tokens/s
17.2
tokens/s
491
ms
pp16+tg1536
62
tokens/s
16.9
tokens/s
310
ms