OpenAI Lifeboat – Proxy OpenAI Code to Llama 70B on Replicate

lifeboat.replicate.dev

6 pointsCharlieholtz3 comments

I understand that LLaMa 70B isn't even at GPT-3's level yet.

If you have paying customers who expect GPT-4-level quality, is using LLaMa 70B to generate responses really better than just telling users that the OpenAI API is temporarily down, from a reputational standpoint?

Llama 70B is at GPT-3.5's level! It's not a perfect comparison, but I built this site to compare the two and Llama 70B won (https://llmboxing.com/leaderboard).

It's not as smart as GPT-4, but you can fine-tune Llama to do things that aren't possible with larger models (https://replicate.com/blog/fine-tune-llama-2).

Clever. That’s pretty neat that they were able to make their API perfectly drop-in compatible.

Reply on news.ycombinator.com