TEST #2919 RESULTS
02/27/2026 - 12:16 PM
ACCELERATOR
MODEL
118
tokens/s
generation
568
ms
time to first token
2164
tokens/s
prompt
767
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
2647
tokens/s
134
tokens/s
394
ms
pp4096+tg256
2154
tokens/s
77.3
tokens/s
1.91
sec
pp2048+tg256
2503
tokens/s
106
tokens/s
828
ms
pp2048+tg768
2499
tokens/s
102
tokens/s
829
ms
pp1024+tg1024
2613
tokens/s
121
tokens/s
399
ms
pp1280+tg3072
2582
tokens/s
95.2
tokens/s
504
ms
pp384+tg1152
2509
tokens/s
137
tokens/s
159
ms
pp64+tg1024
1483
tokens/s
150
tokens/s
49
ms
pp16+tg1536
489
tokens/s
143
tokens/s
39
ms