Updates Hi, everyone, thanks for using stella models. After six months of work, I trained the jasper model on top of the stella model, which is a multimodal model, and it can be ranked 2 in mteb (submitted the results on 2024 12 11, which may need official review https://github.com/embeddings benchmark/results/pull/68). Model link: https://huggingface.co/infgrad/jasper en vision language v1 I'll focus on the technical report, training data and related code, hopefully the tricks I've used will be of some help to you guys! The core training code will be integrated into the rag retrieval library(https://github.com/NLPJCL/RAG Retrieval) in the near future. (Welcome to star) This work was accomplished during my free time, it's a personal hobby. One person's time and energy is limited, and you are welcome to make any contributions! You can also find these models on my homepage. Introduction The models are trained based on Alibaba NLP/gte large en v1.5 and Alibaba NLP/gte Qwen2 1.5B instruct . Thanks for their contributions! We simplify usage of prompts, providing two prompts for most general tasks, one is for s2p, another one is for s2s. Prompt of s2p task(e.g. retrieve task): Prompt of…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy