There’s already project that help with going beyond the context window limitation like https://github.com/jerryjliu/llama_index https://github.com/jerryjliu/lla

There’s already project that help with going beyond the context window limitation like https://github.com/jerryjliu/llama_index https://github.com/jerryjliu/llama_index They also just tweeted this to showcase how it can work with multimodal data too: https://twitter.com/gpt_index/status/1635668512822956032?s=46&t=6ZIqJnQp7YWjtdJnmNgmtg https://twitter.com/gpt_index/status/1635668512822956032?s=4...