Files
Alex Newman 135ac08a9b server: let CLAUDE_MEM_SERVER_MAX_OUTPUT_TOKENS set the generation output cap (#3997)
All three server generation providers already accept maxOutputTokens, but
nothing ever populated it, so every request went out with the constructor
default of 4096. On a model that answers at length the reply is truncated
mid-structure, the observation parser rejects it, and the job settles as a
non-retryable parse_error, so the observation is lost silently.

Unset keeps 4096, so behaviour is unchanged unless the variable is set.
Non-integer and non-positive values are ignored with a warning rather than
sent as NaN.

Co-authored-by: escooterclinic <dvla5@escooterclinic.co.uk>
2026-09-10 21:57:54 -07:00
..