What is the context limit for the File Search API?

Hello,

Excited to try out the File Search API. It seems like smaller size files work well, but when I have tried to upload a 175KB xlsx file with 3 sheets it struggles to import eventually throwing 2 kinds of errors:

- `Upload has already been terminated`
- `Failed to count tokens.`

If I remove 2 sheets it works. If I add back 1 sheet and a few rows to it, it also seems to work. Overall, across the 3 sheets it is about 308K characters. (Mix of Latin and Cyrillic). And I am simply using the default chunking.

It seems like I am hitting some quota somewhere (tokens/context), but cannot see it anywhere mentioned in the doc: https://ai.google.dev/gemini-api/docs/file-search which says it can process files up to 100MB in size.