TEST #1042 RESULTS
06/12/2025 - 11:21 PM
ACCELERATOR
MODEL
21.2
tokens/s
generation
6.41
sec
time to first token
195
tokens/s
prompt
86
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Pro 8P+2E
RAM
32GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:54:49 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
216
tokens/s
22.8
tokens/s
4.78
sec
pp4096+tg256
201
tokens/s
16.5
tokens/s
20.42
sec
pp2048+tg256
212
tokens/s
20.1
tokens/s
9.70
sec
pp2048+tg768
212
tokens/s
19.6
tokens/s
9.70
sec
pp1024+tg1024
219
tokens/s
21.9
tokens/s
4.72
sec
pp1280+tg3072
219
tokens/s
19.0
tokens/s
5.89
sec
pp384+tg1152
214
tokens/s
23.0
tokens/s
1.83
sec
pp64+tg1024
184
tokens/s
24.4
tokens/s
387
ms
pp16+tg1536
74
tokens/s
23.8
tokens/s
256
ms