TEST #4107 RESULTS
08/31/2026 - 8:21 PM
ACCELERATOR
MODEL
203
tokens/s
generation
1.41
sec
time to first token
10746
tokens/s
prompt
1155
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 9 386H (pantherlake)
RAM
60.9GB
OS
Linux
Kernel Release
7.0.0-30-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #30-Ubuntu SMP PREEMPT_DYNAMIC Fri Jul 31 18:22:54 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
13415
tokens/s
220
tokens/s
81
ms
pp4096+tg256
7877
tokens/s
126
tokens/s
530
ms
pp2048+tg256
10451
tokens/s
181
tokens/s
203
ms
pp2048+tg768
11053
tokens/s
170
tokens/s
192
ms
pp1024+tg1024
15073
tokens/s
208
tokens/s
73
ms
pp1280+tg3072
13084
tokens/s
156
tokens/s
103
ms
pp384+tg1152
17008
tokens/s
240
tokens/s
26
ms
pp64+tg1024
8755
tokens/s
270
tokens/s
11
ms
pp16+tg1536
1
tokens/s
253
tokens/s
11.49
sec