tomaarsen (@tomaarsen)

Perplexity AI가 검색(retrieval)용으로 설계된 4개의 오픈-웨이트(state-of-the-art) 다국어 임베딩 모델을 공개했습니다. 대표 모델로 pplx-embed-v1과 pplx-embed-context-v1이 있으며, int8 및 바이너리 임베딩에 특화되어 대규모 검색 문제에 적합하도록 훈련되었다고 알립니다.

https://x.com/tomaarsen/status/2027392224879595949

#perplexity #embeddings #openweights #retrieval #pplxembed

tomaarsen (@tomaarsen) on X

🤗 @perplexity_ai has released 4 open-weights state-of-the-art multilingual embedding models designed for retrieval tasks! pplx-embed-v1 and pplx-embed-context-v1 Specifically trained for int8 and binary embeddings, they'll be viable for massive search problems. Details in 🧵

X (formerly Twitter)