Skip to main content
All documentation
Core Concepts 6 min readUpdated September 2026

Adaptive Context Planner

Fit heterogeneous context into a real model budget without weakening hard constraints.

Create a plan

Choose a model profile, policy, and output-token reserve. The planner generates safe representation candidates, allocates them deterministically, validates the result, and explains every decision.

planner.py

Policies and validation

Balanced, quality-first, cost-first, and latency-first policies change the utility tradeoff, not the trust boundary. Post-plan validation can conservatively fall back to safer candidates.

Optional hosted Sarvam mode

Studio always supports offline planning. Where operators explicitly enable it, the separate hosted route reserves distributed request, token, concurrency, and project-spend quota before calling Sarvam. Paid mode defaults off, BYOK is not stored in the public UI, and estimated metrics are clearly separated from provider-reported usage.

Frequently asked questions