Nicki Gataro

ceoofcapybaras

AI & ML interests

None yet

Recent Activity

Organizations

None yet

ceoofcapybaras's activity

New activity in HighCWu/FLUX.1-dev-4bit 2 months ago

Quantization scripts

13
#1 opened 4 months ago by WaveCut
liked a Space 3 months ago
replied to Xenova's post 3 months ago
view reply

@Ke09876 if the user doesn't have enough storage space, this will fail, moreover, when mobile users run out of memory, Chromium browsers automatically start cleaning things up, removing saved content such as indexedDB. There is something called persistent storage, but it's is a joke, only in Firefox it asks the user for permission, Chrome just on its own decides to grant or reject it, who thought it was a good idea...

replied to Xenova's post 3 months ago
view reply

As a frontend dev, LLMs were not meant for the browsers. You have to download the weights every time you reload the page. It's impressive that they do run well in the browser, but I don't see any practical use cases.

ValueError

2
#2 opened 5 months ago by ceoofcapybaras
New activity in apple/DCLM-7B-8k 5 months ago

Context-length

2
#1 opened 5 months ago by ceoofcapybaras

Load failed: 34

1
#1 opened 7 months ago by ceoofcapybaras
reacted to bwang0911's post with ๐Ÿš€ 8 months ago
view post
Post
2993
In the vector search setup, we normally combine a fast embedding model and an accurate but slow reranker model.

The newly released @jinaai rerankers are small in size and almost as accurate as our base reranker. This means given a time constraint, it can scoring more candidate documents from embedding models and have a better chance to feed LLM the correct context for RAG generation.

These models are available on Huggingface and has been integrated into the latest SentenceTransformers 2.7.0. Check it out!

jinaai/jina-reranker-v1-turbo-en
jinaai/jina-reranker-v1-tiny-en
  • 1 reply
ยท