Answers are grounded in this paper and cite their sources.
Use floor((context_limit−prompt_tokens−output_reserve)/(chunk_payload_tokens+overhead)) to count chunks at 8k, 32k, 128k; output reserve cuts 8k first. Independent technical note mirroring the canonical version: https://x-oss.com/blog/how-many-chunks-fit-at-8k32k128k-a-capacity-formula-then-summarize.php
No takes yet. Share an insight, caveat, or question.
Wei Chen (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: