TEST #2927 RESULTS
02/27/2026 - 1:05 PM
ACCELERATOR
MODEL
37.7
tokens/s
generation
3.11
sec
time to first token
388
tokens/s
prompt
167
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Max 8P+2E
RAM
64GB
OS
Darwin
Kernel Release
25.2.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.2.0: Tue Nov 18 21:09:40 PST 2025; root:xnu-12377.61.12~1/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
448
tokens/s
40.6
tokens/s
2.31
sec
pp4096+tg256
421
tokens/s
30.1
tokens/s
9.76
sec
pp2048+tg256
437
tokens/s
36.0
tokens/s
4.71
sec
pp2048+tg768
441
tokens/s
35.1
tokens/s
4.67
sec
pp1024+tg1024
446
tokens/s
38.4
tokens/s
2.32
sec
pp1280+tg3072
439
tokens/s
33.9
tokens/s
2.94
sec
pp384+tg1152
436
tokens/s
40.9
tokens/s
903
ms
pp64+tg1024
317
tokens/s
42.6
tokens/s
224
ms
pp16+tg1536
104
tokens/s
41.6
tokens/s
178
ms