TEST #3271 RESULTS
04/10/2026 - 9:43 AM
ACCELERATOR
MODEL
169
tokens/s
generation
181
ms
time to first token
9257
tokens/s
prompt
2050
LocalScore
HOW YOU STACK UP
Explore All ResultsLlm 350m Instruct v2 - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-14450HX (alderlake)
RAM
15.7GB
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
11398
tokens/s
173
tokens/s
95
ms
pp4096+tg256
5269
tokens/s
150
tokens/s
784
ms
pp2048+tg256
8725
tokens/s
160
tokens/s
242
ms
pp2048+tg768
8678
tokens/s
168
tokens/s
243
ms
pp1024+tg1024
12871
tokens/s
175
tokens/s
85
ms
pp1280+tg3072
11096
tokens/s
167
tokens/s
122
ms
pp384+tg1152
17345
tokens/s
174
tokens/s
27
ms
pp64+tg1024
6788
tokens/s
177
tokens/s
14
ms
pp16+tg1536
1137
tokens/s
175
tokens/s
19
ms