3 ms·
This is a good start, devxpy! What other features are you planning to build over it? Hit me up if you want to collaborate.
by kubrickslair 3y ago
This is a good start, devxpy! What other features are you planning to build over it? Hit me up if you want to collaborate.
- devxpy 3y agoThe main issue is that its very slow and expensive to browse the internet like this. The LLM will only perform well if you have it do chain of thought reasoning, and that has a latency hit because of a longer generation.