Connection
Base URL, credentials and a reachable server.
Product
Product features · Prompt token budgets
01
Reserve response tokens before assembling a model request. Test a prompt exactly at the remaining limit and stop when the reserve exceeds the context window.
Selection
A model available through the configured endpoint.
Capabilities
Context limits and tool support are determined by the model and server.
Prompt token budgets
● ● ●ModelTrim●
A concrete starting point · Prompt token budgets ×
const availableInput = contextLimit - outputReserve;
if (inputTokens > availableInput) return reduceContext();
return sendWithinBudget();✳Reserve response tokens before assembling a model request.
✓ ChangesReview before applying ↗
Illustrative workspace