Retrieve Message Batch results changedapi/beta/messages/batches/results
Nearest release: v2.1.319, published an hour after this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+2added
Lines−3removed
From line
3,026
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits29to this page, all time
The whole hunk
from line 3026, old and new numbered
/
from line 3026
30263026
30273027 - `max_tokens: optional number or null`
30283028
3029 Bounds the advisor's total output (thinking + text) per call. When the advisor hits this cap, the returned advisor_result or advisor_redacted_result block carries stop_reason='max_tokens', and a truncation note is appended to the advice text the worker model sees (inside the encrypted blob in redacted mode). When set, the server also emits a remaining-tokens budget block in the advisor's prompt so the advisor self-shapes toward the cap. When omitted, the advisor model's default output cap applies and no budget block is emitted.
3029 Bounds the advisor's total output (thinking + text) per call. When the advisor hits this cap, the returned advisor_result or advisor_redacted_result block carries stop_reason='max_tokens', and a truncation note is appended to the advice text the worker model sees (inside the encrypted blob in redacted mode). When set, the API also passes the advisor its remaining-token budget, so the advisor shapes its output toward the cap. When omitted, the advisor model's default output cap applies and the API doesn't pass a budget.
30303030
30313031 minimum: 1024
30323032
from line 3899
38993899 the thinking-block delimiter tokens.
39003900
39013901 Reflects the raw reasoning the model produced, not the (possibly shorter)
3902 summarized thinking text returned in the response body. Computed by
3903 re-tokenizing the raw reasoning text, so it may differ from the model's exact
3902 summarized thinking text returned in the response body. It may differ from the model's exact
39043903 generation count by a small number of tokens. Always ≤ `output_tokens`;
39053904 `output_tokens - thinking_tokens` approximates the non-reasoning output.
39063905
No line in this hunk matches that.