Yibo Peng

add test api transformer_int4_fp16_gpu

1

## Description Add test api transformer_int4_fp16_gpu in all-in-one benchmark. The attached files are results of two apis. [transformer_int4_fp16_gpu-results-2024-04-02.csv](https://github.com/intel-analytics/ipex-llm/files/14834323/transformer_int4_fp16_gpu-results-2024-04-02.csv) [transformer_int4_gpu-results-2024-04-02.csv](https://github.com/intel-analytics/ipex-llm/files/14834325/transformer_int4_gpu-results-2024-04-02.csv)

## Description Add an example of Langchain RAG, which aims to read paper. Currently it can only read text part of the paper. [https://github.com/analytics-zoo/nano/issues/1060](https://github.com/analytics-zoo/nano/issues/1060) Reference: [(https://github.com/langchain-ai/langchain/blob/master/cookbook/Semi_structured_multi_modal_RAG_LLaMA2.ipynb)](https://github.com/langchain-ai/langchain/blob/master/cookbook/Semi_structured_multi_modal_RAG_LLaMA2.ipynb)

## Description init commit of open-webui quickstart TODO - [x] upload pics and a video in repo https://github.com/bigdl-project/bigdl-project.github.io/pull/88 - [x] add the video and pics' links into quickstart

## Description Update 8192 prompt in all-in-one Add cpu_embedding param for linux api