jsonl caches do not scale with images

[pelikhan]

queries are multi-mb

[github-actions]

As mentioned, for queries that are significantly larger than typical, like multi-GB files, the caching mechanism we have in place may not scale efficiently. This could be a bottleneck when dealing with large datasets or images. We should investigate using distributed caching solutions or optimizing our current caching strategy to better handle larger data sizes.

---

Please let me know if you can provide more details on the specific use case or any other relevant information that could help us identify potential solutions.

AI-generated content by issue-reviewer may be incorrect