TEST #3436 RESULTS
05/04/2026 - 3:29 AM
ACCELERATOR
MODEL
28.2
tokens/s
generation
2.10
sec
time to first token
806
tokens/s
prompt
221
LocalScore
HOW YOU STACK UP
Explore All ResultsPhi 4 Mini Instruct - Q6_K
SYSTEM
CPU
AMD Ryzen 7 5700G with Radeon Graphics (znver3)
RAM
15GB
OS
Linux
Kernel Release
6.19.13-arch1-1
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Tue, 21 Apr 2026 23:38:22 +0000
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
926
tokens/s
35.9
tokens/s
1.13
sec
pp4096+tg256
471
tokens/s
4.1
tokens/s
8.94
sec
pp2048+tg256
721
tokens/s
25.2
tokens/s
2.88
sec
pp2048+tg768
722
tokens/s
23.4
tokens/s
2.87
sec
pp1024+tg1024
938
tokens/s
30.7
tokens/s
1.12
sec
pp1280+tg3072
894
tokens/s
15.2
tokens/s
1.46
sec
pp384+tg1152
1110
tokens/s
37.4
tokens/s
368
ms
pp64+tg1024
958
tokens/s
42.8
tokens/s
86
ms
pp16+tg1536
511
tokens/s
38.9
tokens/s
51
ms