Y
Hacker News
new
|
ask
|
show
|
jobs
by
bbotond
1078 days ago
What I don’t understand is why is an API needed to create embeddings. Isn’t this something that could be done locally?
4 comments
pantulis
1078 days ago
You would need to have a local copy of the GPT model, which are not exactly OpenAI's plans.
link
jerrygenser
1078 days ago
For embeddings, you can use smaller transformers/llms or sentence2vec and often get good enough results.
You don't need very large models to generate usable embeddings.
link
pantulis
1078 days ago
You are correct, I assumed parent was referring to specific embeddings generated by OpenAI LLMs.
link
thorum
1078 days ago
It’s cheaper to use OpenAI. If you have your own compute, sentence-transformers is just as good for most use cases.
link
teaearlgraycold
1078 days ago
Yes. The best public embedding model is decent, but I expect it’s objectively worse than the best model from OpenAI.
link
merpnderp
1078 days ago
Sure, but I don't know of any models you can get local access to that work nearly as well.
link