TEST #2933 RESULTS
02/27/2026 - 1:33 PM
ACCELERATOR
MODEL
117
tokens/s
generation
570
ms
time to first token
2163
tokens/s
prompt
764
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 10P+4E
RAM
48GB
OS
Darwin
Kernel Release
25.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.1.0: Mon Oct 20 19:34:05 PDT 2025; root:xnu-12377.41.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2655
tokens/s
133
tokens/s
393
ms
pp4096+tg256
2178
tokens/s
76.9
tokens/s
1.89
sec
pp2048+tg256
2472
tokens/s
106
tokens/s
838
ms
pp2048+tg768
2477
tokens/s
101
tokens/s
836
ms
pp1024+tg1024
2592
tokens/s
120
tokens/s
403
ms
pp1280+tg3072
2560
tokens/s
94.8
tokens/s
508
ms
pp384+tg1152
2458
tokens/s
136
tokens/s
163
ms
pp64+tg1024
1698
tokens/s
149
tokens/s
44
ms
pp16+tg1536
377
tokens/s
141
tokens/s
48
ms