TEST #944 RESULTS
05/25/2025 - 9:52 AM
ACCELERATOR
MODEL
30.7
tokens/s
generation
4.33
sec
time to first token
291
tokens/s
prompt
127
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 8P+4E
RAM
24GB
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 19:53:27 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
321
tokens/s
33.0
tokens/s
3.23
sec
pp4096+tg256
295
tokens/s
23.8
tokens/s
13.91
sec
pp2048+tg256
315
tokens/s
29.1
tokens/s
6.54
sec
pp2048+tg768
316
tokens/s
28.3
tokens/s
6.52
sec
pp1024+tg1024
325
tokens/s
31.2
tokens/s
3.18
sec
pp1280+tg3072
320
tokens/s
27.2
tokens/s
4.04
sec
pp384+tg1152
334
tokens/s
33.4
tokens/s
1.18
sec
pp64+tg1024
283
tokens/s
34.9
tokens/s
253
ms
pp16+tg1536
114
tokens/s
35.1
tokens/s
166
ms