Multi-Modal Retrieval: GPT text embedding and CLIP image embeddinggithub.com/run-llama2 pointsomarfarooq3 years ago