TEST #2354 RESULTS
01/25/2026 - 6:54 PM
ACCELERATOR
MODEL
91.8
tokens/s
generation
314
ms
time to first token
3860
tokens/s
prompt
1041
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-10310U CPU @ 1.70GHz (skylake)
RAM
62.1GB
OS
Linux
Kernel Release
6.14.0-37-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #37~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Nov 20 10:25:38 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3876
tokens/s
93.9
tokens/s
274
ms
pp4096+tg256
4100
tokens/s
84.6
tokens/s
1.01
sec
pp2048+tg256
4847
tokens/s
94.0
tokens/s
433
ms
pp2048+tg768
4867
tokens/s
88.1
tokens/s
431
ms
pp1024+tg1024
4736
tokens/s
91.6
tokens/s
227
ms
pp1280+tg3072
4518
tokens/s
85.4
tokens/s
294
ms
pp384+tg1152
4568
tokens/s
95.2
tokens/s
94
ms
pp64+tg1024
2054
tokens/s
97.8
tokens/s
42
ms
pp16+tg1536
1176
tokens/s
96.0
tokens/s
23
ms