Isn‘t clip superseeded by multimodal llms?
In this program CLIP is being used to create embeddings. A multimodal LLM does something very similar. In this case the language model is not needed because the embeddings are being used to search directly.
Show HN: Sisi – Semantic Image Search CLI tool, locally without third party APIs
41–44 of 44 posts
Re: Show HN: Sisi – Semantic Image Search CLI tool, locally without third party APIs
#42I have made similar android app for semantic image search, works offline too, still gathering feedback and polishing UI, but it works, if you are brave enough here is it https://drive.google.com/file/d/1tE0cY6umj5h5zCY_Jvaou1M8sCf...
Why yes, I'll download a 695MB APK file from an internet stranger.
Yes, the size is 99% 2 models weights required to run inference offline, there no way around it.
Re: Show HN: Sisi – Semantic Image Search CLI tool, locally without third party APIs
#43using same app rclip: https://github.com/yurijmikhalevich/rclip
Re: Show HN: Sisi – Semantic Image Search CLI tool, locally without third party APIs
#44In russian, "sisi" is a variation of "tits". Is there a job/services that confirm that branding is appropriate across different languages? Seems like a non trivial problem to solve.
> Seems like a non trivial problem to solve. Took me 5 minutes to land this GPT prompt. https://chatgpt.com/share/66e84c0c-a92c-800a-b452-255d6fe942... Results: - Chinese (Simplified) 四四 (sì sì) – sounds like "four-four", which can be associated with bad luck due to the number four in Chinese culture - Arabic "Sisi" is a common nickname, also associated with Egypt's President Abdel Fattah el-Sisi - Russian Сиси (sisi…
Tried with "hui" - for ChatGPT this word "has no specific meaning and is safe for use in any language".