ai-forever

/

charllama-2.6B

like 7

Hi all,

I an kind a new in this field so I am sorry if I am asking something obvious:-)
I am running this model on a laptop with Intel i7-11800H CPU and NVIDIA GeForce RTX 3070 Laptop GPU.
It seems like that CPU has twice performance as compared to GPU:

У бу́рных чу́вств неи́стовый коне́ц,
Когда́ на се́рдце ра́на кровото́чит.
Когда́ в груди́ стучи́т т
RESUME:
Used cuda:0 device (NVIDIA GeForce RTX 3070 Laptop GPU)
Generated 99 chars
Generation took 52.18339514732361 seconds
Speed was 1.8971552103979483 char per second

У бу́рных чу́вств неи́стовый коне́ц,
Когда́ они́ нахо́дят вы́ход в сло́ве.
Поэ́т в стиха́х - не то́
RESUME:
Used cpu device (11th Gen Intel(R) Core(TM) i7-11800H @ 2.30GHz (2.30 GHz))
Generated 99 chars
Generation took 29.71364974975586 seconds
Speed was 3.3318020786326805 char per second

Just wondering if this is an expected behaviour?