DeepSeek launched a free, open-source large language model in late December, claiming it was developed in just two months at a cost of under $6 million.
if, on a modern gaming pc, you can get breakneck speeds of 5 tokens per second, then actually inference is quite energy intensive too. 5 per second of anything is very slow
if, on a modern gaming pc, you can get breakneck speeds of 5 tokens per second, then actually inference is quite energy intensive too. 5 per second of anything is very slow