TEST #3568 RESULTS
05/18/2026 - 3:41 PM
ACCELERATOR
MODEL
196
tokens/s
generation
188
ms
time to first token
8034
tokens/s
prompt
2028
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-14650HX (alderlake)
RAM
15.3GB
OS
Linux
Kernel Release
6.18.7-76061807-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202601231045~1769703228~24.04~cb87b5b SMP PREEMPT_DYNAMIC Thu J
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
10435
tokens/s
206
tokens/s
103
ms
pp4096+tg256
5508
tokens/s
175
tokens/s
749
ms
pp2048+tg256
8016
tokens/s
192
tokens/s
261
ms
pp2048+tg768
7978
tokens/s
189
tokens/s
262
ms
pp1024+tg1024
10259
tokens/s
197
tokens/s
105
ms
pp1280+tg3072
9527
tokens/s
185
tokens/s
139
ms
pp384+tg1152
14229
tokens/s
201
tokens/s
32
ms
pp64+tg1024
5667
tokens/s
209
tokens/s
16
ms
pp16+tg1536
685
tokens/s
206
tokens/s
28
ms