PreparedModelCall

CORE · interface · Type contract

Request-bound, single-use transport snapshot.

Repeated measurements do not prepare or spend calls.

Import

TypeScript

import type { PreparedModelCall } from '@jangts/agent-core/model';

JavaScript

/** @typedef {import('@jangts/agent-core/model').PreparedModelCall} PreparedModelCall */

Other public entry points:@jangts/agent-core, @jangts/agent-core/model, @jangts/agent-core/types, @jangts/agent-core/structs.

Definition

interface PreparedModelCall<Result>

Extends or implements: PreparedModelTransport<Result>.

Type parameters

Type parameter	Constraint	Default type
Result	—	unknown

Field structure

Field	Type	Required	Description
modelId	string | undefined	Optional	Read-only. Model binding the call was prepared with.
stageId	string	Required	Read-only. See the type definition and usage context.
transportId	string	Required	Read-only. See the type definition and usage context.
maxOutputTokens	number	Required	Read-only. See the type definition and usage context.
remainingCalls	number	Required	Read-only. Calls still available; a live value, so later spends are visible.
measure	(input: string | readonly ModelMessage[] | BuiltContext) => ModelInputMeasurement	Required	Preflight measurement of one input shape. Never marks the call used and never charges the budget.
options	Pick<ModelOptions, "functions" | "json" | "toolChoice" | "parallelToolCalls" | "temperature">	Required	Read-only. See the type definition and usage context.
schemaChars	number	Required	Read-only. See the type definition and usage context.
instruction	string | undefined	Optional	Read-only. See the type definition and usage context.
createResponse	() => ModelResponseCollector<Result>	Required	See the type definition and usage context.

Member index

Member	Purpose
measure	Preflight measurement of one input shape. Never marks the call used and never charges the budget.
createResponse	See the overload below for full inputs and outputs.

Member details

measure

Preflight measurement of one input shape.

Never marks the call used and never charges the budget.

PreparedModelCall.measure(input: string | readonly ModelMessage[] | BuiltContext): ModelInputMeasurement

Preflight measurement of one input shape.

Never marks the call used and never charges the budget.

Parameter	Type	Passing	Description
input	string | readonly ModelMessage[] | BuiltContext	Required	—

Optional shapes of Parameter input: choose one shape; required fields for each branch are listed separately.

Branch 1

string

Branch 2

readonly ModelMessage[]

Branch 3

Field	Type	Required	Description
messages	ModelMessage[]	Required	Messages in final model order.
fragmentUsage	ContextFragmentUsage[]	Required	Per-fragment accounting in the same order as the fragments were merged.
inputChars	number	Required	Total counted characters across messages and usage.
maxInputChars	number	Required	Budget the assembly was validated against; Infinity means unbounded.

Returns: ModelInputMeasurement.

Return value structure

Field	Type	Required	Description
messageChars	number	Required	Read-only. Characters of the message content.
instructionChars	number	Required	Read-only. Characters of the instruction text.
schemaChars	number	Required	Read-only. Characters of the compiled capability schemas.
transportChars	number	Required	Read-only. Characters the transport itself contributes (for example framing).
inputChars	number	Required	Read-only. Sum of the character components above.
maxInputChars	number	Required	Read-only. Character budget the input was compared against.
maxSchemaChars	number	Required	Read-only. Schema budget the input was compared against.
maxOutputTokens	number	Required	Read-only. Output tokens reserved for this call, which reduces the token input ceiling.
counterId	string | undefined	Optional	Read-only. Tokenizer identity; present only when a token budget applies.
inputTokens	number | undefined	Optional	Read-only. Measured input tokens.
inputTokenLimit	number | undefined	Optional	Read-only. Token ceiling after reserving output and the safety margin.
withinBudget	boolean	Required	Read-only. True when every applicable character and token budget accepted the input.

createResponse

PreparedModelCall.createResponse(): ModelResponseCollector<Result>

No call parameters.

Returns: ModelResponseCollector<Result>.

Return value structure

Field	Type	Required	Description
call	{ id: string; name: string; argumentChars: number; } | undefined	Optional	Read-only. Native function call that carried the selection, when the response used one.
calls	readonly { id: string; name: string; argumentChars: number; toolId?: string; }[] | undefined	Optional	Read-only. All native calls carried by a session response; omitted by single-selection collectors.
outputChars	number	Required	Read-only. Output characters accumulated so far, for output-budget accounting.
push	(part: ModelPart) => void	Required	Appends one streamed part; parts arrive in provider order.
finish	() => Result	Required	Finalizes the response. Throws when the accumulated content is not a valid selection.

Examples

• Actual request measurement and same-call preparation snapshot: Use prepare only when context must be handled ahead of time; multiple measurements and the final send reuse the same transport preparation.

Ordinary calls still only use invoke

Beginner · Node.js

TypeScript

import { ModelRunner } from '@jangts/agent-core/model';
const provider = {
  async *stream(messages: readonly { content: string }[]) {
    yield '收到：' + messages.at(-1)!.content;
  },
};
const runner = new ModelRunner({ provider });
console.log((await runner.invoke('你好')).value);

JavaScript

import { ModelRunner } from '@jangts/agent-core/model';
const provider = {
    async *stream(messages) {
        yield '收到：' + messages.at(-1).content;
    },
};
const runner = new ModelRunner({ provider });
console.log((await runner.invoke('你好')).value);

Prepare once only when measurement is needed

Configuration · Node.js

TypeScript

import { ModelRunner } from '@jangts/agent-core/model';
const provider = {
  async *stream(messages: readonly { content: string }[]) {
    yield '收到：' + messages.at(-1)!.content;
  },
};
const runner = new ModelRunner({ provider, maxInputChars: 1000 });
const prepared = runner.prepare({ stageId: 'reply', maxOutputTokens: 400 });
const input = '你好，解释这次服务状态';
console.log(prepared.measure(input));
const result = await runner.invoke({ input, prepared });
console.log(result.value);

JavaScript

import { ModelRunner } from '@jangts/agent-core/model';
const provider = {
    async *stream(messages) {
        yield '收到：' + messages.at(-1).content;
    },
};
const runner = new ModelRunner({ provider, maxInputChars: 1000 });
const prepared = runner.prepare({ stageId: 'reply', maxOutputTokens: 400 });
const input = '你好，解释这次服务状态';
console.log(prepared.measure(input));
const result = await runner.invoke({ input, prepared });
console.log(result.value);

Related policy source

• Same-call preparation and real input measurement

Related definitions

• BuiltContext
• ModelInputMeasurement
• ModelMessage
• ModelOptions
• ModelResponseCollector
• PreparedModelTransport