Evidence report
Findings
Ranked recommendations from the latest independent analysis.
Latest report
Jun 28, 2026 – Jul 28, 2026 · Jul 28, 2026
The strongest opportunities are in model routing and repeated prompt context. Provider usage and runtime traces agree that a large low-complexity request segment uses the most capable model, while repository evidence shows a stable instruction prefix repeated across calls. Retry behavior is worth tightening, but its measured impact is less certain.
Ranked recommendations
Ordered by evidence and practical priority
Route low-complexity extraction traffic to a smaller model
Structured extraction accounts for 38% of calls and has narrow outputs, stable latency, and low observed retry rates.
Cache the stable policy and schema prefix
A large prompt prefix is identical across extraction calls and is assembled before request-specific context.
Bound duplicate retries after client timeouts
A small request cohort is retried after client timeouts even when the original inference later completes.
Other opportunities
Candidates outside the ranked set
Tighten retrieval count for support answers
Candidate is plausible, but repository and runtime signals do not yet identify a safe context threshold.