CtrlK
BlogDocsLog inGet started
Tessl Logo

tessl/maven-io-quarkiverse-langchain4j--quarkus-langchain4j-watsonx

Quarkus extension for integrating IBM watsonx.ai foundation models with LangChain4j. Provides chat models, generation models, streaming models, embedding models, and scoring models for IBM watsonx.ai. Includes comprehensive configuration options, support for tool/function calling, text extraction from documents in Cloud Object Storage, and experimental built-in services for Google search, weather, and web crawling. Designed for enterprise Java applications using the Quarkus framework with built-in dependency injection and native compilation support.

Overview
Eval results
Files

text-extraction.mddocs/

Text Extraction

Extract text content from documents (PDF, images) stored in IBM Cloud Object Storage (COS). Supports multiple output formats including JSON, Markdown, HTML, and plain text. Useful for preprocessing documents before RAG ingestion or analysis.

Capabilities

Text Extraction Service

Service for extracting text from documents in IBM Cloud Object Storage.

public class TextExtraction {
    public TextExtraction(
        Reference documentReference,
        Reference resultReference,
        String projectId,
        String spaceId,
        String version,
        COSRestApi cosClient,
        WatsonxRestApi watsonxClient
    );

    // Synchronous methods (start extraction and wait for result)
    public String extractAndFetch(String absolutePath);
    public String extractAndFetch(String absolutePath, Parameters parameters);
    public String uploadExtractAndFetch(File file);
    public String uploadExtractAndFetch(File file, Parameters parameters);
    public String uploadExtractAndFetch(InputStream is, String fileName);
    public String uploadExtractAndFetch(InputStream is, String fileName, Parameters parameters);

    // Asynchronous methods (start extraction only)
    public String startExtraction(String absolutePath);
    public String startExtraction(String absolutePath, Parameters parameters);
    public String uploadAndStartExtraction(File file);
    public String uploadAndStartExtraction(File file, Parameters parameters);
    public String uploadAndStartExtraction(InputStream is, String fileName);
    public String uploadAndStartExtraction(InputStream is, String fileName, Parameters parameters);

    // Status and cleanup
    public TextExtractionResponse checkExtractionStatus(String id);
    public void deleteFile(String bucketName, String path);
    public void deleteFile(String bucketName, String path, Duration timeout);
}

Supported Input Formats:

  • PDF (.pdf)
  • GIF (.gif)
  • JPG/JPEG (.jpg, .jpeg)
  • PNG (.png)
  • TIFF (.tif, .tiff)

Supported Output Formats:

  • JSON - Structured JSON with text and metadata
  • MARKDOWN - Markdown formatted text
  • HTML - HTML formatted text
  • PLAIN_TEXT - Plain text without formatting
  • PAGE_IMAGES - Extracted page images

Basic Usage

Synchronous Extraction (Upload and Extract):

import io.quarkiverse.langchain4j.watsonx.runtime.TextExtraction;
import jakarta.inject.Inject;
import java.io.File;

@ApplicationScoped
public class DocumentProcessor {
    @Inject
    TextExtraction textExtraction;

    public String extractPdf(File pdfFile) {
        // Upload file, extract text, and return result
        String extractedText = textExtraction.uploadExtractAndFetch(pdfFile);
        return extractedText;
    }
}

Asynchronous Extraction:

import io.quarkiverse.langchain4j.watsonx.bean.TextExtractionResponse;

public class AsyncDocumentProcessor {
    @Inject
    TextExtraction textExtraction;

    public String startExtraction(File pdfFile) {
        // Start extraction job and return job ID
        String jobId = textExtraction.uploadAndStartExtraction(pdfFile);
        return jobId;
    }

    public String checkStatus(String jobId) {
        TextExtractionResponse status = textExtraction.checkExtractionStatus(jobId);

        switch (status.metadata().status()) {
            case COMPLETED:
                // Fetch result from COS
                String result = fetchResultFromCOS(status.entity().resultsReference());
                return result;
            case FAILED:
                throw new RuntimeException("Extraction failed: " + status.entity().serviceError());
            case RUNNING:
            case QUEUED:
                return "Still processing...";
            default:
                return "Unknown status";
        }
    }
}

Text Extraction Reference

Reference to Cloud Object Storage bucket for input documents or output results.

public record Reference(String connection, String bucket) {
    public Reference(String connection);  // bucket = null
    public Reference(String connection, String bucket);
}

Example:

// Reference with both connection and bucket
TextExtraction.Reference documentRef = new TextExtraction.Reference(
    "cos-connection-id",
    "input-documents"
);

// Reference with connection only (bucket specified elsewhere)
TextExtraction.Reference resultsRef = new TextExtraction.Reference(
    "cos-connection-id"
);

Text Extraction Parameters

Configuration parameters for text extraction operations.

public class Parameters {
    public static Builder builder();

    public Duration getTimeout();
    public String getOutputFileName();
    public List<Type> getTypes();
    public Optional<Reference> getDocumentReference();
    public Optional<Reference> getResultsReference();
    public Optional<Boolean> getRemoveUploadedFile();
    public Optional<Boolean> getRemoveOutputFile();
    public Mode getMode();
    public OCR getOcr();
    public Boolean getAutoRotationCorrection();
    public EmbeddedImages getEmbeddedImages();
    public Integer getDpi();
    public Boolean getOutputTokensAndBbox();

    public static class Builder {
        public Builder timeout(Duration timeout);
        public Builder outputFileName(String outputFileName);
        public Builder types(List<Type> types);
        public Builder types(Type type);
        public Builder types(Type... types);
        public Builder documentReference(Reference documentReference);
        public Builder resultsReference(Reference resultsReference);
        public Builder removeUploadedFile(Boolean removeUploadedFile);
        public Builder removeOutputFile(Boolean removeOutputFile);
        public Builder mode(Mode mode);
        public Builder ocr(OCR ocr);
        public Builder autoRotationCorrection(Boolean autoRotationCorrection);
        public Builder embeddedImages(EmbeddedImages embeddedImages);
        public Builder dpi(Integer dpi);
        public Builder outputTokensAndBbox(Boolean outputTokensAndBbox);
        public Parameters build();
    }
}

Parameter Details:

  • timeout (Duration): Maximum time to wait for extraction completion

    • Default: 60 seconds
    • Only applies to synchronous methods (extractAndFetch, uploadExtractAndFetch)

  • outputFileName (String): Name of output file in results bucket

    • If not specified, auto-generated name is used

  • types (List<Type>): Output format types

    • Values: Type.JSON, Type.MARKDOWN, Type.HTML, Type.PLAIN_TEXT, Type.PAGE_IMAGES
    • Can specify multiple types
    • Default: Type.MARKDOWN

  • documentReference (Reference): Override default document bucket reference

  • resultsReference (Reference): Override default results bucket reference

  • removeUploadedFile (Boolean): Delete uploaded file after extraction

    • Default: false
    • Only applies when using upload methods

  • removeOutputFile (Boolean): Delete output file after retrieval

    • Default: false
    • Only applies to synchronous methods

  • mode (Mode): Processing mode for extraction

    • Values depend on document type and model capabilities

  • ocr (OCR): OCR configuration for image-based text extraction

    • Values depend on OCR capabilities

  • autoRotationCorrection (Boolean): Automatically correct page rotation

    • Default: false
    • Useful for scanned documents

  • embeddedImages (EmbeddedImages): How to handle embedded images

    • Values depend on output format

  • dpi (Integer): Target DPI for image processing

    • Higher DPI = better quality but slower processing

  • outputTokensAndBbox (Boolean): Include token positions and bounding boxes

    • Default: false
    • Useful for advanced document analysis

Examples:

// Simple extraction to markdown
TextExtraction.Parameters params = TextExtraction.Parameters.builder()
    .types(Type.MARKDOWN)
    .timeout(Duration.ofMinutes(2))
    .build();

// Multiple output formats
TextExtraction.Parameters multiFormat = TextExtraction.Parameters.builder()
    .types(Type.MARKDOWN, Type.JSON, Type.PLAIN_TEXT)
    .outputFileName("extracted-content")
    .build();

// Advanced OCR with auto-rotation
TextExtraction.Parameters ocrParams = TextExtraction.Parameters.builder()
    .types(Type.MARKDOWN)
    .autoRotationCorrection(true)
    .dpi(300)
    .outputTokensAndBbox(true)
    .build();

// Clean up files after extraction
TextExtraction.Parameters cleanupParams = TextExtraction.Parameters.builder()
    .types(Type.PLAIN_TEXT)
    .removeUploadedFile(true)
    .removeOutputFile(true)
    .build();

Extraction Methods

Synchronous Extraction from COS

Extract text from files already in Cloud Object Storage.

// Extract from existing file in COS
String result = textExtraction.extractAndFetch("/path/to/document.pdf");

// With custom parameters
TextExtraction.Parameters params = TextExtraction.Parameters.builder()
    .types(Type.MARKDOWN)
    .timeout(Duration.ofMinutes(3))
    .build();

String result = textExtraction.extractAndFetch("/path/to/document.pdf", params);

Synchronous Extraction with Upload

Upload local file and extract in one operation.

import java.io.File;
import java.io.FileInputStream;

// Upload file and extract
File pdfFile = new File("/local/path/document.pdf");
String result = textExtraction.uploadExtractAndFetch(pdfFile);

// Upload from input stream
try (FileInputStream fis = new FileInputStream(pdfFile)) {
    String result = textExtraction.uploadExtractAndFetch(
        fis,
        "document.pdf"
    );
}

// With custom parameters
TextExtraction.Parameters params = TextExtraction.Parameters.builder()
    .types(Type.JSON, Type.MARKDOWN)
    .removeUploadedFile(true)
    .removeOutputFile(true)
    .build();

String result = textExtraction.uploadExtractAndFetch(pdfFile, params);

Asynchronous Extraction from COS

Start extraction job and check status separately.

// Start extraction job
String jobId = textExtraction.startExtraction("/path/to/document.pdf");

// Check status later
TextExtractionResponse status = textExtraction.checkExtractionStatus(jobId);

if (status.metadata().status() == TextExtractionResponse.Status.COMPLETED) {
    // Retrieve result from COS
    String resultPath = status.entity().resultsReference().location();
    String result = fetchFromCOS(resultPath);
}

Asynchronous Extraction with Upload

Upload and start extraction job.

import java.io.File;

// Upload and start extraction
File pdfFile = new File("/local/path/document.pdf");
String jobId = textExtraction.uploadAndStartExtraction(pdfFile);

// Poll for completion
TextExtractionResponse status;
do {
    Thread.sleep(5000);  // Wait 5 seconds
    status = textExtraction.checkExtractionStatus(jobId);
} while (status.metadata().status() == TextExtractionResponse.Status.RUNNING ||
         status.metadata().status() == TextExtractionResponse.Status.QUEUED);

if (status.metadata().status() == TextExtractionResponse.Status.COMPLETED) {
    System.out.println("Extraction completed!");
}

Cleanup Operations

Delete files from Cloud Object Storage.

// Delete file from bucket
textExtraction.deleteFile("bucket-name", "/path/to/file.pdf");

// Delete with custom timeout
textExtraction.deleteFile(
    "bucket-name",
    "/path/to/file.pdf",
    Duration.ofSeconds(30)
);

// Clean up after extraction
String jobId = textExtraction.uploadAndStartExtraction(file);
TextExtractionResponse status = textExtraction.checkExtractionStatus(jobId);

if (status.metadata().status() == TextExtractionResponse.Status.COMPLETED) {
    // Delete input and output files
    textExtraction.deleteFile(inputBucket, inputPath);
    textExtraction.deleteFile(outputBucket, outputPath);
}

Dependency Injection with Quarkus

Configure and inject TextExtraction service.

Configuration:

quarkus.langchain4j.watsonx.base-url=https://us-south.ml.cloud.ibm.com
quarkus.langchain4j.watsonx.api-key=your-api-key
quarkus.langchain4j.watsonx.project-id=your-project-id

# Text extraction configuration
quarkus.langchain4j.watsonx.text-extraction.base-url=https://s3.us-south.cloud-object-storage.appdomain.cloud
quarkus.langchain4j.watsonx.text-extraction.document-reference.connection=input-connection-id
quarkus.langchain4j.watsonx.text-extraction.document-reference.bucket-name=input-documents
quarkus.langchain4j.watsonx.text-extraction.results-reference.connection=output-connection-id
quarkus.langchain4j.watsonx.text-extraction.results-reference.bucket-name=output-results
quarkus.langchain4j.watsonx.text-extraction.log-requests=false
quarkus.langchain4j.watsonx.text-extraction.log-responses=false

Injection:

import io.quarkiverse.langchain4j.watsonx.runtime.TextExtraction;
import jakarta.inject.Inject;

@ApplicationScoped
public class DocumentService {
    @Inject
    TextExtraction textExtraction;

    public String extractDocument(File file) {
        return textExtraction.uploadExtractAndFetch(file);
    }
}

Types

Text Extraction Response

public record TextExtractionResponse(
    Metadata metadata,
    Entity entity
) {
    public record Metadata(
        String id,
        String href,
        Status status,
        Instant createdAt,
        Instant completedAt
    ) {}

    public record Entity(
        ResultsReference resultsReference,
        ServiceError serviceError
    ) {}

    public record ResultsReference(
        String location,
        String type,
        String connection,
        String bucket
    ) {}

    public record ServiceError(
        String code,
        String message
    ) {}

    public enum Status {
        QUEUED,      // Extraction job queued
        RUNNING,     // Extraction in progress
        COMPLETED,   // Extraction completed successfully
        FAILED       // Extraction failed
    }
}

Usage Example:

String jobId = textExtraction.startExtraction("/path/to/doc.pdf");
TextExtractionResponse response = textExtraction.checkExtractionStatus(jobId);

System.out.println("Job ID: " + response.metadata().id());
System.out.println("Status: " + response.metadata().status());
System.out.println("Created at: " + response.metadata().createdAt());

if (response.metadata().status() == TextExtractionResponse.Status.COMPLETED) {
    System.out.println("Result location: " + response.entity().resultsReference().location());
} else if (response.metadata().status() == TextExtractionResponse.Status.FAILED) {
    System.out.println("Error: " + response.entity().serviceError().message());
}

Text Extraction Request

Internal request bean:

public class TextExtractionRequest {
    public static Builder builder();

    public static class Builder {
        public Builder documentReference(DocumentReference documentReference);
        public Builder resultsReference(ResultsReference resultsReference);
        public Builder parameters(ExtractionParameters parameters);
        public Builder projectId(String projectId);
        public Builder spaceId(String spaceId);
        public TextExtractionRequest build();
    }
}

Output Types

public enum Type {
    JSON,         // Structured JSON output
    MD,           // Markdown output (same as MARKDOWN)
    MARKDOWN,     // Markdown output
    HTML,         // HTML output
    PLAIN_TEXT,   // Plain text output
    PAGE_IMAGES   // Page images output
}

Processing Enums

// Processing mode
public enum Mode {
    // Mode values depend on document type and capabilities
}

// OCR configuration
public enum OCR {
    // OCR values depend on OCR capabilities
}

// Embedded images handling
public enum EmbeddedImages {
    // Values depend on output format
}

RAG Integration

Extract and Ingest Documents

import dev.langchain4j.data.document.Document;
import dev.langchain4j.data.document.DocumentSplitter;
import dev.langchain4j.data.document.splitter.DocumentSplitters;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.EmbeddingStoreIngestor;
import java.io.File;

@ApplicationScoped
public class DocumentIngestor {
    @Inject
    TextExtraction textExtraction;

    @Inject
    EmbeddingModel embeddingModel;

    @Inject
    EmbeddingStore<TextSegment> embeddingStore;

    public void ingestPdf(File pdfFile) {
        // Extract text from PDF
        String extractedText = textExtraction.uploadExtractAndFetch(pdfFile);

        // Create document
        Document document = Document.from(
            extractedText,
            Metadata.from("source", pdfFile.getName())
        );

        // Split into chunks
        DocumentSplitter splitter = DocumentSplitters.recursive(500, 50);
        List<TextSegment> segments = splitter.split(document);

        // Embed and store
        EmbeddingStoreIngestor ingestor = EmbeddingStoreIngestor.builder()
            .embeddingModel(embeddingModel)
            .embeddingStore(embeddingStore)
            .build();

        ingestor.ingest(segments);
    }
}

Batch Document Processing

import java.util.List;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

@ApplicationScoped
public class BatchDocumentProcessor {
    @Inject
    TextExtraction textExtraction;

    private final ExecutorService executor = Executors.newFixedThreadPool(10);

    public List<String> processDocumentsBatch(List<File> files) {
        // Start all extraction jobs asynchronously
        List<CompletableFuture<String>> futures = files.stream()
            .map(file -> CompletableFuture.supplyAsync(() -> {
                // Upload and start extraction
                String jobId = textExtraction.uploadAndStartExtraction(file);

                // Poll for completion
                TextExtractionResponse status;
                do {
                    Thread.sleep(5000);
                    status = textExtraction.checkExtractionStatus(jobId);
                } while (status.metadata().status() == TextExtractionResponse.Status.RUNNING ||
                         status.metadata().status() == TextExtractionResponse.Status.QUEUED);

                if (status.metadata().status() == TextExtractionResponse.Status.COMPLETED) {
                    // Fetch result
                    return fetchResultFromCOS(status.entity().resultsReference());
                } else {
                    throw new RuntimeException("Extraction failed: " +
                        status.entity().serviceError().message());
                }
            }, executor))
            .toList();

        // Wait for all to complete
        return futures.stream()
            .map(CompletableFuture::join)
            .toList();
    }
}

Extract with Metadata Preservation

@ApplicationScoped
public class MetadataPreservingExtractor {
    @Inject
    TextExtraction textExtraction;

    public Document extractWithMetadata(File file, Map<String, Object> additionalMetadata) {
        // Extract text
        TextExtraction.Parameters params = TextExtraction.Parameters.builder()
            .types(Type.JSON)  // JSON includes structural metadata
            .outputTokensAndBbox(true)  // Include token positions
            .build();

        String jsonResult = textExtraction.uploadExtractAndFetch(file, params);

        // Parse JSON to extract metadata
        JsonObject json = parseJson(jsonResult);
        String text = extractTextFromJson(json);
        Map<String, Object> extractedMetadata = extractMetadataFromJson(json);

        // Combine with additional metadata
        Metadata metadata = Metadata.from(extractedMetadata);
        additionalMetadata.forEach(metadata::put);

        return Document.from(text, metadata);
    }
}

Best Practices

Timeout Configuration

// Short timeout for small documents
TextExtraction.Parameters quickParams = TextExtraction.Parameters.builder()
    .timeout(Duration.ofSeconds(30))
    .build();

// Longer timeout for large documents
TextExtraction.Parameters largeParams = TextExtraction.Parameters.builder()
    .timeout(Duration.ofMinutes(5))
    .build();

Resource Cleanup

// Always clean up temporary files
TextExtraction.Parameters cleanupParams = TextExtraction.Parameters.builder()
    .removeUploadedFile(true)
    .removeOutputFile(true)
    .build();

String result = textExtraction.uploadExtractAndFetch(file, cleanupParams);
// Input and output files are automatically deleted

Error Handling

import io.quarkiverse.langchain4j.watsonx.exception.TextExtractionException;
import io.quarkiverse.langchain4j.watsonx.exception.COSException;

public String extractSafely(File file) {
    try {
        return textExtraction.uploadExtractAndFetch(file);
    } catch (TextExtractionException e) {
        System.err.println("Extraction failed: " + e.getMessage());
        // Handle extraction-specific errors
        return null;
    } catch (COSException e) {
        System.err.println("COS error: " + e.getMessage() +
            " (Status: " + e.statusCode() + ")");
        // Handle COS-specific errors
        return null;
    } catch (Exception e) {
        System.err.println("Unexpected error: " + e.getMessage());
        return null;
    }
}

Asynchronous Processing for Large Batches

// For large document batches, use asynchronous methods
public void processLargeBatch(List<File> files) {
    // Start all extractions
    Map<String, File> jobIdToFile = new HashMap<>();
    for (File file : files) {
        String jobId = textExtraction.uploadAndStartExtraction(file);
        jobIdToFile.put(jobId, file);
    }

    // Poll periodically
    ScheduledExecutorService scheduler = Executors.newScheduledThreadPool(1);
    scheduler.scheduleAtFixedRate(() -> {
        for (Map.Entry<String, File> entry : jobIdToFile.entrySet()) {
            String jobId = entry.getKey();
            TextExtractionResponse status = textExtraction.checkExtractionStatus(jobId);

            if (status.metadata().status() == TextExtractionResponse.Status.COMPLETED) {
                // Process completed extraction
                String result = fetchResultFromCOS(status.entity().resultsReference());
                processResult(result, entry.getValue());
                jobIdToFile.remove(jobId);
            } else if (status.metadata().status() == TextExtractionResponse.Status.FAILED) {
                // Handle failed extraction
                handleFailure(entry.getValue(), status.entity().serviceError());
                jobIdToFile.remove(jobId);
            }
        }

        // Stop scheduler when all done
        if (jobIdToFile.isEmpty()) {
            scheduler.shutdown();
        }
    }, 0, 10, TimeUnit.SECONDS);
}

Output Format Selection

// Plain text for simple RAG ingestion
TextExtraction.Parameters plainParams = TextExtraction.Parameters.builder()
    .types(Type.PLAIN_TEXT)
    .build();

// Markdown for structured content
TextExtraction.Parameters mdParams = TextExtraction.Parameters.builder()
    .types(Type.MARKDOWN)
    .build();

// JSON for detailed analysis
TextExtraction.Parameters jsonParams = TextExtraction.Parameters.builder()
    .types(Type.JSON)
    .outputTokensAndBbox(true)
    .build();

// Multiple formats for different use cases
TextExtraction.Parameters multiParams = TextExtraction.Parameters.builder()
    .types(Type.MARKDOWN, Type.JSON, Type.PLAIN_TEXT)
    .build();

Types

TextExtractionRequest

Bean classes for text extraction API requests and responses.

public record TextExtractionRequest(
    String spaceId,
    String projectId,
    TextExtractionDataReference documentReference,
    TextExtractionDataReference resultsReference,
    TextExtractionParameters parameters
) {
    public static Builder builder();

    public static record CosDataLocation(String fileName, String bucket) {}

    public static record CosDataConnection(String id) {}

    public static record TextExtractionDataReference(
        String type,
        CosDataConnection connection,
        CosDataLocation location
    ) {
        public static TextExtractionDataReference of(String connectionId, String fileName);
        public static TextExtractionDataReference of(String connectionId, String fileName, String bucket);
    }

    public static record TextExtractionParameters(
        Mode mode,
        OCR ocr,
        List<Type> output,
        String outputFileName,
        Boolean autoRotationCorrection,
        EmbeddedImages embeddedImages,
        Integer dpi,
        Boolean outputTokensAndBbox
    ) {}

    public static class Builder {
        public Builder spaceId(String spaceId);
        public Builder projectId(String projectId);
        public Builder documentReference(TextExtractionDataReference documentReference);
        public Builder resultsReference(TextExtractionDataReference resultsReference);
        public Builder parameters(TextExtractionParameters parameters);
        public TextExtractionRequest build();
    }

    public static enum Type {
        JSON,
        HTML,
        MD,
        PLAIN_TEXT,
        PAGE_IMAGES
    }

    public static enum Mode {
        STANDARD,
        HIGH_QUALITY
    }

    public static enum OCR {
        DISABLED,
        ENABLED,
        FORCED
    }

    public static enum EmbeddedImages {
        DISABLED,
        ENABLED_PLACEHOLDER,
        ENABLED_TEXT,
        ENABLED_VERBALIZATION,
        ENABLED_VERBALIZATION_ALL
    }
}

Enum Descriptions:

Type (Output Formats):

  • JSON: Structured JSON with detailed layout information
  • HTML: HTML representation preserving document structure
  • MD: Markdown format for easy text processing
  • PLAIN_TEXT: Plain text without formatting
  • PAGE_IMAGES: Individual page images

Mode:

  • STANDARD: Standard extraction quality (faster)
  • HIGH_QUALITY: High quality extraction (slower, more accurate)

OCR:

  • DISABLED: Disable OCR entirely
  • ENABLED: Enable OCR for images when needed
  • FORCED: Force OCR for all pages

EmbeddedImages:

  • DISABLED: Ignore embedded images
  • ENABLED_PLACEHOLDER: Replace images with placeholders
  • ENABLED_TEXT: Extract text from images
  • ENABLED_VERBALIZATION: Generate verbal descriptions of images
  • ENABLED_VERBALIZATION_ALL: Generate detailed descriptions for all images

Install with Tessl CLI

npx tessl i tessl/maven-io-quarkiverse-langchain4j--quarkus-langchain4j-watsonx@1.7.0

docs

builtin-services.md

chat-models.md

configuration.md

embedding-scoring.md

exceptions.md

generation-models.md

index.md

request-parameters.md

text-extraction.md

tile.json