TEST #1575 RESULTS
09/30/2025 - 7:07 AM
ACCELERATOR
MODEL
22.7
tokens/s
generation
6.04
sec
time to first token
207
tokens/s
prompt
92
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Pro 8P+2E
RAM
16GB
OS
Darwin
Kernel Release
25.0.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.0.0: Mon Aug 25 21:17:51 PDT 2025; root:xnu-12377.1.9~3/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
231
tokens/s
24.5
tokens/s
4.48
sec
pp4096+tg256
214
tokens/s
17.6
tokens/s
19.24
sec
pp2048+tg256
226
tokens/s
21.5
tokens/s
9.11
sec
pp2048+tg768
226
tokens/s
21.0
tokens/s
9.11
sec
pp1024+tg1024
230
tokens/s
23.2
tokens/s
4.50
sec
pp1280+tg3072
229
tokens/s
20.2
tokens/s
5.63
sec
pp384+tg1152
230
tokens/s
24.8
tokens/s
1.71
sec
pp64+tg1024
194
tokens/s
25.9
tokens/s
366
ms
pp16+tg1536
80
tokens/s
25.3
tokens/s
237
ms