TEST #4042 RESULTS
08/17/2026 - 9:56 PM
ACCELERATOR
MODEL
31.0
tokens/s
generation
4.15
sec
time to first token
303
tokens/s
prompt
131
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 8P+4E
RAM
48GB
OS
Darwin
Kernel Release
25.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.5.0: Tue Jun 9 22:28:34 PDT 2026; root:xnu-12377.121.10~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
338
tokens/s
33.3
tokens/s
3.06
sec
pp4096+tg256
311
tokens/s
24.1
tokens/s
13.20
sec
pp2048+tg256
330
tokens/s
29.5
tokens/s
6.24
sec
pp2048+tg768
328
tokens/s
28.7
tokens/s
6.29
sec
pp1024+tg1024
337
tokens/s
31.5
tokens/s
3.07
sec
pp1280+tg3072
331
tokens/s
27.6
tokens/s
3.90
sec
pp384+tg1152
334
tokens/s
34.0
tokens/s
1.18
sec
pp64+tg1024
295
tokens/s
35.5
tokens/s
244
ms
pp16+tg1536
126
tokens/s
34.4
tokens/s
153
ms