TEST #1379 RESULTS

08/15/2025 - 1:08 AM

32.6

tokens/s

generation

3.64

sec

time to first token

332

tokens/s

prompt

144

LocalScore

HOW YOU STACK UP
Explore All Results

Granite 3.2 8b Instruct - Q4_K - Medium

SYSTEM
CPU
Apple M1 Max 8P+2E
RAM
64GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Jul 14 11:30:29 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
386
tokens/s
35.3
tokens/s
2.68
sec
pp4096+tg256
359
tokens/s
25.5
tokens/s
11.44
sec
pp2048+tg256
377
tokens/s
31.2
tokens/s
5.47
sec
pp2048+tg768
378
tokens/s
29.8
tokens/s
5.45
sec
pp1024+tg1024
369
tokens/s
33.2
tokens/s
2.81
sec
pp1280+tg3072
378
tokens/s
29.0
tokens/s
3.41
sec
pp384+tg1152
375
tokens/s
35.7
tokens/s
1.05
sec
pp64+tg1024
279
tokens/s
37.4
tokens/s
255
ms
pp16+tg1536
90
tokens/s
36.4
tokens/s
203
ms