TEST #1992 RESULTS
12/20/2025 - 4:35 PM
ACCELERATOR
MODEL
276
tokens/s
generation
99
ms
time to first token
12950
tokens/s
prompt
3310
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-8700 CPU @ 3.20GHz (skylake)
RAM
15.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
17373
tokens/s
287
tokens/s
63
ms
pp4096+tg256
12188
tokens/s
246
tokens/s
340
ms
pp2048+tg256
15523
tokens/s
271
tokens/s
136
ms
pp2048+tg768
15351
tokens/s
270
tokens/s
137
ms
pp1024+tg1024
17499
tokens/s
281
tokens/s
62
ms
pp1280+tg3072
16398
tokens/s
262
tokens/s
82
ms
pp384+tg1152
17594
tokens/s
283
tokens/s
25
ms
pp64+tg1024
3770
tokens/s
295
tokens/s
20
ms
pp16+tg1536
854
tokens/s
291
tokens/s
23
ms