TEST #1300 RESULTS
08/03/2025 - 3:47 PM
ACCELERATOR
MODEL
61.9
tokens/s
generation
537
ms
time to first token
2495
tokens/s
prompt
660
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2680 v4 @ 2.40GHz (broadwell)
RAM
94.1GB
OS
Linux
Kernel Release
6.14.8-2-bpo12-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 6.14.8-2~bpo12+1 (2025-07-22T13:15Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2931
tokens/s
63.9
tokens/s
365
ms
pp4096+tg256
2249
tokens/s
55.9
tokens/s
1.84
sec
pp2048+tg256
2653
tokens/s
61.1
tokens/s
789
ms
pp2048+tg768
2648
tokens/s
60.5
tokens/s
790
ms
pp1024+tg1024
2933
tokens/s
62.8
tokens/s
365
ms
pp1280+tg3072
2823
tokens/s
59.3
tokens/s
469
ms
pp384+tg1152
3248
tokens/s
64.1
tokens/s
134
ms
pp64+tg1024
2163
tokens/s
65.3
tokens/s
45
ms
pp16+tg1536
805
tokens/s
64.5
tokens/s
35
ms