@molecule/api-ocr-llm

Language-model OCR provider for molecule.dev — extracts text from images through any vision-capable @molecule/api-ai model

Auto-generated, AI-first package reference for the molecule.dev ecosystem. It is written to be read by coding agents as much as by people, and is generated from this package's source — edit src/index.ts JSDoc, not this file.

Type

provider

Installation

npm install @molecule/api-ocr-llm @molecule/api-ai @molecule/api-ocr

API

Interfaces

LlmOcrConfig

Options for {@link createProvider}. ai falls back to the bonded ai provider; every other field has a prompt-level default.

interface LlmOcrConfig {
  /**
   * The AI provider to recognize with. Defaults to the provider bonded as
   * `ai` (`@molecule/api-ai` `requireProvider()`), which must support image
   * input (check the model catalog's `supportsVision`).
   */
  ai?: AIProvider
  /** The model id to pass to the provider. Omit to use the provider's default. */
  model?: string
  /** Default language hint when a `recognize` call passes none. */
  language?: string
  /**
   * Extra instructions, e.g. "This is a receipt; transcribe every line item".
   * Appended to the system prompt.
   */
  instructions?: string
  /** Sampling temperature. Defaults to 0 — transcription is not a creative task. */
  temperature?: number
  /** Output token ceiling. Defaults to a length derived from the image size. */
  maxOutputTokens?: number
}

Functions

systemPrompt(instructions)

The system prompt for one recognition pass.

function systemPrompt(instructions?: string): string

• instructions — Extra caller instructions, appended last.

Returns: The system prompt.

userPrompt(language)

The text part of the user message (the image block is sent alongside it).

function userPrompt(language?: string): string

• language — Optional language hint.

Returns: The user-side instruction.

createProvider(config)

Creates a language-model OCR provider.

function createProvider(config?: LlmOcrConfig): OcrProvider

• config — Which AI provider/model to use and how to prompt it.

Returns: An OcrProvider backed by a vision language model.

Constants

provider

The provider implementation.

const provider: OcrProvider

Core Interface

Implements @molecule/api-ocr interface.

Bond Wiring

Setup function to register this provider with the core interface:

import { setProvider } from '@molecule/api-ocr'
import { provider } from '@molecule/api-ocr-llm'

export function setupOcrLlm(): void {
  setProvider(provider)
}

Injection Notes

Requirements

Peer dependencies:

• @molecule/api-ai >=1.0.0
• @molecule/api-ocr >=1.0.0

Runtime Dependencies

• @molecule/api-ai
• @molecule/api-ocr