Most of those inference providers are American, and China is actually at a disadvantage here because of export restrictions - US companies are using newer and more efficient chips.

▲

amazingamazing 6 days ago | parent [-]

If it’s newer and efficient then why is the api more expensive?

▲

veber-alex 5 days ago | parent [-]

Price is set based on what people are willing to pay not based on actual costs.

	▲	amazingamazing 5 days ago \| parent [-]
		I’d believe that if they didn’t lower limits