TEST #3888 RESULTS
07/10/2026 - 1:31 PM
ACCELERATOR
MODEL
97.8
tokens/s
generation
189
ms
time to first token
7018
tokens/s
prompt
1538
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i9-13900K (alderlake)
RAM
63.8GB
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
9371
tokens/s
97.4
tokens/s
119
ms
pp4096+tg256
6588
tokens/s
94.5
tokens/s
632
ms
pp2048+tg256
8291
tokens/s
98.7
tokens/s
258
ms
pp2048+tg768
8228
tokens/s
98.9
tokens/s
260
ms
pp1024+tg1024
9441
tokens/s
97.7
tokens/s
118
ms
pp1280+tg3072
6333
tokens/s
96.2
tokens/s
213
ms
pp384+tg1152
9473
tokens/s
98.1
tokens/s
51
ms
pp64+tg1024
4228
tokens/s
99.8
tokens/s
24
ms
pp16+tg1536
1210
tokens/s
98.4
tokens/s
23
ms