Skip to main content

Module prepared_request

Module prepared_request 

Source
Expand description

Prepared conditioning request snapshot and exact-token admission gating.

Enforces:

  1. Dual limits: both max_bytes and input_tokens verified against the exact final rendered stream.
  2. Separate output reservation: actual_input_tokens + reserved_output_tokens <= supported_context_tokens.
  3. Full metadata snapshot: binds profile ID, version, evidence revision, access scope, model instance, adapter hash, and rendered token digest into the prepared request.
  4. Generation of a canonical InferenceCacheKey compatible with the Tier-1 prefix cache.

Structsยง

PreparedConditioningSnapshot
Snapshot of a fully validated and admitted conditioning request ready for execution.