TEST #3207 RESULTS
03/28/2026 - 9:37 PM
ACCELERATOR
MODEL
58.8
tokens/s
generation
1.72
sec
time to first token
698
tokens/s
prompt
288
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
128GB
OS
Darwin
Kernel Release
25.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.3.0: Wed Jan 28 20:51:28 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
823
tokens/s
64.1
tokens/s
1.26
sec
pp4096+tg256
758
tokens/s
44.9
tokens/s
5.42
sec
pp2048+tg256
800
tokens/s
55.6
tokens/s
2.58
sec
pp2048+tg768
800
tokens/s
54.1
tokens/s
2.58
sec
pp1024+tg1024
813
tokens/s
60.3
tokens/s
1.28
sec
pp1280+tg3072
813
tokens/s
51.8
tokens/s
1.59
sec
pp384+tg1152
785
tokens/s
64.7
tokens/s
503
ms
pp64+tg1024
507
tokens/s
67.7
tokens/s
141
ms
pp16+tg1536
181
tokens/s
66.1
tokens/s
102
ms