TEST #1131 RESULTS
06/27/2025 - 1:32 AM
ACCELERATOR
MODEL
189
tokens/s
generation
169
ms
time to first token
9573
tokens/s
prompt
2204
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i9-14900KS (alderlake)
RAM
244.2GB
OS
Linux
Kernel Release
6.14.0-22-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #22-Ubuntu SMP PREEMPT_DYNAMIC Wed May 21 15:01:51 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
12048
tokens/s
195
tokens/s
90
ms
pp4096+tg256
5949
tokens/s
168
tokens/s
695
ms
pp2048+tg256
8829
tokens/s
187
tokens/s
237
ms
pp2048+tg768
8784
tokens/s
184
tokens/s
239
ms
pp1024+tg1024
12019
tokens/s
193
tokens/s
91
ms
pp1280+tg3072
11178
tokens/s
178
tokens/s
120
ms
pp384+tg1152
16939
tokens/s
197
tokens/s
27
ms
pp64+tg1024
7962
tokens/s
201
tokens/s
13
ms
pp16+tg1536
2447
tokens/s
198
tokens/s
11
ms