mixpeek 0.81.1

package/dist/models/parameters.d.ts

@@ -0,0 +1,728 @@
+/**
+ * Mixpeek API
+ * This is the Mixpeek API, providing access to various endpoints for data processing and retrieval.
+ *
+ * The version of the OpenAPI document: 0.81.1
+ * Contact: info@mixpeek.com
+ *
+ * NOTE: This class is auto generated by OpenAPI Generator (https://openapi-generator.tech).
+ * https://openapi-generator.tech
+ * Do not edit the class manually.
+ */
+import type { ChunkStrategy } from './chunk-strategy';
+import type { CrawlMode } from './crawl-mode';
+import type { DocumentIdStrategy } from './document-id-strategy';
+import type { GenerationConfig } from './generation-config';
+import type { RenderStrategy } from './render-strategy';
+import type { ResponseShape3 } from './response-shape3';
+import type { SplitMethod } from './split-method';
+import type { TextSplitStrategy } from './text-split-strategy';
+/**
+ * Parameters for the feature extractor. Each extractor type has specific parameters. See the schema for your chosen extractor (e.g., MultimodalExtractorParams for multimodal_extractor).
+ * @export
+ * @interface Parameters
+ */
+export interface Parameters {
+    /**
+     * Custom plugin extractor type (plugin name)
+     * @type {string}
+     * @memberof Parameters
+     */
+    'extractor_type': string;
+    /**
+     * Source content type. Use \'youtube\' to resolve YouTube URLs to caption text before embedding. Default: \'text\' (plain text input).
+     * @type {string}
+     * @memberof Parameters
+     */
+    'source_type'?: ParametersSourceTypeEnum;
+    /**
+     * Strategy for splitting text into multiple documents.
+     * @type {TextSplitStrategy}
+     * @memberof Parameters
+     */
+    'split_by'?: TextSplitStrategy;
+    /**
+     * Target size for each chunk (in units of chunk_strategy).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'chunk_size'?: number;
+    /**
+     * Overlap between chunks to preserve context.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'chunk_overlap'?: number;
+    /**
+     * Length of each transcript segment in seconds (for time_segments split strategy). Shorter segments give more precise search results but more documents.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'segment_length_seconds'?: number;
+    /**
+     * Preferred language code for YouTube captions (when source_type=\'youtube\').
+     * @type {string}
+     * @memberof Parameters
+     */
+    'language'?: string;
+    /**
+     * Extract auto-captions or manual subtitles from YouTube videos (when source_type=\'youtube\'). Falls back to video description if False.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'extract_captions'?: boolean;
+    /**
+     *
+     * @type {ResponseShape3}
+     * @memberof Parameters
+     */
+    'response_shape'?: ResponseShape3;
+    /**
+     * LLM provider for structured extraction: openai, google, anthropic
+     * @type {string}
+     * @memberof Parameters
+     */
+    'llm_provider'?: string;
+    /**
+     * LLM model for structured extraction.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'llm_model'?: string;
+    /**
+     * API key for LLM operations (BYOK - Bring Your Own Key). Supports: - Direct key: \'sk-proj-abc123...\' - Secret reference: \'{{SECRET.openai_api_key}}\'  When using secret reference, the key is loaded from your organization\'s secrets vault at runtime. Store secrets via POST /v1/organizations/secrets.  If not provided, uses Mixpeek\'s default API keys.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'llm_api_key'?: string;
+    /**
+     * The PRIMARY control for video splitting strategy. This determines which splitting method is used.
+     * @type {SplitMethod}
+     * @memberof Parameters
+     */
+    'split_method'?: SplitMethod;
+    /**
+     * The prompt to use for description generation.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'description_prompt'?: string;
+    /**
+     * Interval in seconds for \'time\' splitting. Used when split_method=\'time\'.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'time_split_interval'?: number;
+    /**
+     * The decibel level below which audio is considered silent. Used when split_method=\'silence\'. Recommended value: -40 (auto-applied if not specified). Lower values (e.g., -50) detect more silence, higher values (e.g., -30) detect less.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'silence_db_threshold'?: number;
+    /**
+     * Scene detection sensitivity (0.0-1.0).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'scene_detection_threshold'?: number;
+    /**
+     * Whether to run transcription on video segments.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_transcription'?: boolean;
+    /**
+     * The language of the transcription. Used when run_transcription is True.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'transcription_language'?: string;
+    /**
+     * Whether to generate descriptions for video segments. OPTIMIZED: Defaults to False as descriptions add 1-2 minutes. Enable only when needed.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_video_description'?: boolean;
+    /**
+     * Whether to generate embeddings for transcriptions. Useful for semantic search over spoken content.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_transcription_embedding'?: boolean;
+    /**
+     * Whether to generate multimodal embeddings for all content types (video/image/gif/text). Uses Google Vertex AI to create unified 1408D embeddings in a shared semantic space. Useful for cross-modal semantic search across all media types.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_multimodal_embedding'?: boolean;
+    /**
+     * Whether to run OCR to extract text from video frames. OPTIMIZED: Defaults to False as OCR adds significant processing time. Enable only when text extraction from video is required.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_ocr'?: boolean;
+    /**
+     * The sensitivity of the scene detection.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'sensitivity'?: string;
+    /**
+     * Whether to generate thumbnail images.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'enable_thumbnails'?: boolean;
+    /**
+     * Use CDN for thumbnail delivery.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'use_cdn'?: boolean;
+    /**
+     *
+     * @type {GenerationConfig}
+     * @memberof Parameters
+     */
+    'generation_config'?: GenerationConfig;
+    /**
+     * SCRFD model for face detection. \'scrfd_500m\': Fastest (2-3ms). \'scrfd_2.5g\': Balanced (5-7ms), recommended. \'scrfd_10g\': Highest accuracy (10-15ms).
+     * @type {string}
+     * @memberof Parameters
+     */
+    'detection_model'?: ParametersDetectionModelEnum;
+    /**
+     * Minimum face size in pixels to detect. 20px: Balanced. 40px: Higher quality. 10px: Maximum recall.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'min_face_size'?: number;
+    /**
+     * Confidence threshold for face detection (0.0-1.0).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'detection_threshold'?: number;
+    /**
+     * Maximum number of faces to process per image. None: Process all.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'max_faces_per_image'?: number;
+    /**
+     * L2-normalize embeddings to unit vectors (recommended).
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'normalize_embeddings'?: boolean;
+    /**
+     * Compute quality scores (blur, size, landmarks). Adds ~5ms per face.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'enable_quality_scoring'?: boolean;
+    /**
+     * Minimum quality score to index faces. None: Index all faces. 0.5: Moderate filtering. 0.7: High quality only.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'quality_threshold'?: number;
+    /**
+     * Maximum video length in seconds. 60: Default. 10: Recommended for retrieval. 300: Maximum (extraction only).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'max_video_length'?: number;
+    /**
+     * Frames per second to sample from video. 1.0: One frame per second (recommended).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'video_sampling_fps'?: number;
+    /**
+     * Remove duplicate faces across video frames (extraction only). Reduces 90-95% redundancy. NOT used in retrieval.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'video_deduplication'?: boolean;
+    /**
+     * Cosine similarity threshold for deduplication. 0.8: Conservative (default).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'video_deduplication_threshold'?: number;
+    /**
+     * \'per_face\': One document per face (recommended). \'per_image\': One doc per image with faces array.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'output_mode'?: ParametersOutputModeEnum;
+    /**
+     * Include aligned 112×112 face crops as base64. Adds ~5KB per face.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'include_face_crops'?: boolean;
+    /**
+     * Include resized source frame/image as base64 thumbnail (~15-30KB per face). Used for display with bounding box overlay.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'include_source_frame_thumbnail'?: boolean;
+    /**
+     * Store bbox, landmarks, detection scores. Recommended for debugging.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'store_detection_metadata'?: boolean;
+    /**
+     * Enable ML-based layout detection to find ALL document elements (text, images, tables, figures). When enabled, uses the configured layout_detector to detect and extract both text regions AND non-text elements (scanned images, figures, charts) as separate documents. **Recommended for**: Scanned documents, image-heavy PDFs, mixed content documents. **When disabled**: Falls back to text-only extraction (faster but misses images). Default: True (detects all elements including images).
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'use_layout_detection'?: boolean;
+    /**
+     * Layout detection engine to use when use_layout_detection=True. \'pymupdf\': Fast, rule-based detection using PyMuPDF heuristics (~15 pages/sec). \'docling\': SOTA ML-based detection using IBM Docling with DiT model (~3-8 sec/doc). **Docling advantages**: Better semantic type detection (section_header vs paragraph), true table structure extraction (rows/cols), more accurate figure detection. **PyMuPDF advantages**: Much faster, lower memory usage, simpler dependencies. Default: \'pymupdf\' for speed. Use \'docling\' for accuracy-critical applications.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'layout_detector'?: ParametersLayoutDetectorEnum;
+    /**
+     * Maximum vertical gap (in points) between lines to be grouped in same block. Increase for looser grouping, decrease for tighter blocks. Default 15pt works well for standard documents.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'vertical_threshold'?: number;
+    /**
+     * Maximum horizontal distance (in points) for overlap detection. Affects column detection and block merging. Increase for wider columns, decrease for narrow layouts.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'horizontal_threshold'?: number;
+    /**
+     * Minimum text length (characters) to keep a block. Blocks with less text are filtered out. Helps remove noise and tiny fragments.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'min_text_length'?: number;
+    /**
+     * Base confidence score for embedded (native) text. Penalties are subtracted for OCR artifacts, encoding issues, etc.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'base_confidence'?: number;
+    /**
+     * Confidence threshold below which VLM correction is triggered. Blocks with confidence < this value get sent to VLM for correction. Only applies when use_vlm_correction=True.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'min_confidence_for_vlm'?: number;
+    /**
+     * Enable VLM (Vision Language Model) correction for low-confidence blocks. Uses Gemini/GPT-4V to correct OCR errors by analyzing the page image. Significantly slower (~1 page/sec) but improves accuracy for degraded docs.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'use_vlm_correction'?: boolean;
+    /**
+     * Skip VLM correction entirely for maximum throughput (~15 pages/sec). Overrides use_vlm_correction. Use when speed is more important than accuracy.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'fast_mode'?: boolean;
+    /**
+     * VLM provider: \'google\' (Gemini API) or \'vllm\' (local GPU with Qwen2.5-VL).
+     * @type {string}
+     * @memberof Parameters
+     */
+    'vlm_provider'?: ParametersVlmProviderEnum;
+    /**
+     * VLM model. For google: \'gemini-2.5-flash\'. For vllm: \'Qwen/Qwen2.5-VL-7B-Instruct\'.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'vlm_model'?: string;
+    /**
+     * Generate E5 text embeddings (1024D) for transcripts and text.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_text_embedding'?: boolean;
+    /**
+     * DPI for page rendering (used for VLM correction). 72: Fast, lower quality. 150: Balanced (recommended). 300: High quality, slower.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'render_dpi'?: number;
+    /**
+     * Generate thumbnail images for each learning unit.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'generate_thumbnails'?: boolean;
+    /**
+     * Thumbnail generation mode. \'full_page\': Low-res thumbnail of entire page. \'segment\': Cropped thumbnail of just the block\'s bounding box. \'both\': Generate both types (recommended for flexibility).
+     * @type {string}
+     * @memberof Parameters
+     */
+    'thumbnail_mode'?: string;
+    /**
+     * DPI for thumbnail generation. Lower DPI = smaller files. 72: Standard web quality. 36: Very small thumbnails.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'thumbnail_dpi'?: number;
+    /**
+     * HuggingFace model name for sentiment classification
+     * @type {string}
+     * @memberof Parameters
+     */
+    'model_name'?: string;
+    /**
+     * Maximum token length
+     * @type {number}
+     * @memberof Parameters
+     */
+    'max_length'?: number;
+    /**
+     * Inference batch size
+     * @type {number}
+     * @memberof Parameters
+     */
+    'batch_size'?: number;
+    /**
+     * Return scores for all classes, not just top
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'return_all_scores'?: boolean;
+    /**
+     * Generate E5 embeddings for semantic retrieval alongside classification. Uses the internal E5 embedding service for 1024-dimensional vectors.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'embed'?: boolean;
+    /**
+     * Maximum link depth to crawl. 0=seed page only, 1=seed+direct links, etc. Default: 2. Max: 10.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'max_depth'?: number;
+    /**
+     * Maximum pages to crawl. Default: 50. Max: 500.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'max_pages'?: number;
+    /**
+     * Maximum total time for crawling in seconds. Default: 300 (5 minutes). Increase for large sites with many pages. Max: 3600 (1 hour).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'crawl_timeout'?: number;
+    /**
+     * Crawl strategy. DETERMINISTIC: BFS all links (predictable). SEMANTIC: LLM-guided, prioritizes relevant pages (requires crawl_goal).
+     * @type {CrawlMode}
+     * @memberof Parameters
+     */
+    'crawl_mode'?: CrawlMode;
+    /**
+     * Goal for semantic crawling. Only used when crawl_mode=SEMANTIC. Example: \'Find all S3 API documentation and examples\'
+     * @type {string}
+     * @memberof Parameters
+     */
+    'crawl_goal'?: string;
+    /**
+     * How to render pages. AUTO (default): tries static, falls back to JS. STATIC: fast HTTP fetch. JAVASCRIPT: Playwright browser for SPAs.
+     * @type {RenderStrategy}
+     * @memberof Parameters
+     */
+    'render_strategy'?: RenderStrategy;
+    /**
+     * Regex patterns for URLs to include. Example: [\'/docs/\', \'/api/\']
+     * @type {Array<string>}
+     * @memberof Parameters
+     */
+    'include_patterns'?: Array<string>;
+    /**
+     * Regex patterns for URLs to exclude. Example: [\'/blog/\', \'\\.pdf$\']
+     * @type {Array<string>}
+     * @memberof Parameters
+     */
+    'exclude_patterns'?: Array<string>;
+    /**
+     * How to split page content. NONE: one chunk per page. SENTENCES/PARAGRAPHS: semantic boundaries. WORDS/CHARACTERS: fixed size chunks.
+     * @type {ChunkStrategy}
+     * @memberof Parameters
+     */
+    'chunk_strategy'?: ChunkStrategy;
+    /**
+     * How to generate document IDs. URL (default): stable across re-crawls. POSITION: order-based. CONTENT: deduplicates identical content.
+     * @type {DocumentIdStrategy}
+     * @memberof Parameters
+     */
+    'document_id_strategy'?: DocumentIdStrategy;
+    /**
+     * Generate E5 embeddings for text content.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'generate_text_embeddings'?: boolean;
+    /**
+     * Generate Jina code embeddings for code blocks.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'generate_code_embeddings'?: boolean;
+    /**
+     * Generate SigLIP embeddings for images/figures.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'generate_image_embeddings'?: boolean;
+    /**
+     * Generate DINOv2 visual structure embeddings for layout comparison.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'generate_structure_embeddings'?: boolean;
+    /**
+     * Maximum retry attempts for failed HTTP requests. Uses exponential backoff with jitter. Default: 3.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'max_retries'?: number;
+    /**
+     * Base delay in seconds for retry backoff. Actual delay = base * 2^attempt + jitter. Default: 1.0.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'retry_base_delay'?: number;
+    /**
+     * Maximum delay in seconds between retries. Default: 30.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'retry_max_delay'?: number;
+    /**
+     * Respect Retry-After header from 429/503 responses. If False, uses exponential backoff instead. Default: True.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'respect_retry_after'?: boolean;
+    /**
+     * List of proxy URLs for rotation. Supports formats: \'http://host:port\', \'http://user:pass@host:port\', \'socks5://host:port\'. Proxies rotate on errors or every N requests.
+     * @type {Array<string>}
+     * @memberof Parameters
+     */
+    'proxies'?: Array<string>;
+    /**
+     * Rotate to next proxy when request fails. Default: True.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'rotate_proxy_on_error'?: boolean;
+    /**
+     * Rotate proxy every N requests (0 = disabled). Useful for avoiding IP-based rate limits. Default: 0 (disabled).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'rotate_proxy_every_n_requests'?: number;
+    /**
+     * Captcha solving service provider: \'2captcha\', \'anti-captcha\', \'capsolver\'. If not set, captcha pages are skipped gracefully.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'captcha_service_provider'?: string;
+    /**
+     * API key for captcha solving service. Supports secret reference: \'{{SECRET.captcha_api_key}}\'. Required if captcha_service_provider is set.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'captcha_service_api_key'?: string;
+    /**
+     * Detect captcha challenges (Cloudflare, reCAPTCHA, hCaptcha). If detected and no solver configured, page is skipped. Default: True.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'detect_captcha'?: boolean;
+    /**
+     * Persist cookies across requests within a crawl session. Useful for sites requiring authentication. Default: True.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'persist_cookies'?: boolean;
+    /**
+     * Custom HTTP headers to include in all requests. Example: {\'Authorization\': \'Bearer token\', \'X-Custom\': \'value\'}
+     * @type {{ [key: string]: string; }}
+     * @memberof Parameters
+     */
+    'custom_headers'?: {
+        [key: string]: string;
+    };
+    /**
+     * Delay in seconds between consecutive requests. Useful for polite crawling and avoiding rate limits. Default: 0 (no delay).
+     * @type {number}
+     * @memberof Parameters
+     */
+    'delay_between_requests'?: number;
+    /**
+     * Target duration for video segments in milliseconds.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'target_segment_duration_ms'?: number;
+    /**
+     * Minimum duration for video segments in milliseconds.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'min_segment_duration_ms'?: number;
+    /**
+     * Video segmentation method: \'scene\', \'srt\', or \'time\'.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'segmentation_method'?: ParametersSegmentationMethodEnum;
+    /**
+     * Use Whisper ASR for transcription instead of SRT subtitles.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'use_whisper_asr'?: boolean;
+    /**
+     * Expand each segment into multiple granular documents.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'expand_to_granular_docs'?: boolean;
+    /**
+     * Number of frames to OCR per video segment.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'ocr_frames_per_segment'?: number;
+    /**
+     * How to extract PDF content: \'per_page\' or \'per_element\'.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'pdf_extraction_mode'?: ParametersPdfExtractionModeEnum;
+    /**
+     * DPI for rendering PDF pages/elements as images.
+     * @type {number}
+     * @memberof Parameters
+     */
+    'pdf_render_dpi'?: number;
+    /**
+     * Whether to detect code blocks in PDF text.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'detect_code_in_pdf'?: boolean;
+    /**
+     * Whether to segment code files into individual functions.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'segment_functions'?: boolean;
+    /**
+     * Programming languages to extract from code archives.
+     * @type {Array<string>}
+     * @memberof Parameters
+     */
+    'supported_languages'?: Array<string>;
+    /**
+     * Generate Jina Code embeddings (768D) for code snippets.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_code_embedding'?: boolean;
+    /**
+     * Generate SigLIP visual embeddings (768D) for video frames.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_visual_embedding'?: boolean;
+    /**
+     * Generate DINOv2 visual structure embeddings (768D) for layout comparison.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_structure_embedding'?: boolean;
+    /**
+     * Content type preset for visual embedding strategy.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'visual_embedding_use_case'?: ParametersVisualEmbeddingUseCaseEnum;
+    /**
+     * Run OCR on video frames to extract on-screen text.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'extract_screen_text'?: boolean;
+    /**
+     * Run VLM on video frame thumbnails to extract structured fields: frame_type, page_context, ui_labels, workflow_steps, config_options. Enables drift detection and UI comparison use cases.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'run_vlm_frame_analysis'?: boolean;
+    /**
+     * Use Gemini to generate summaries and enhance descriptions.
+     * @type {boolean}
+     * @memberof Parameters
+     */
+    'enrich_with_llm'?: boolean;
+    /**
+     * Prompt for LLM enrichment when enrich_with_llm=True.
+     * @type {string}
+     * @memberof Parameters
+     */
+    'llm_prompt'?: string;
+}
+export declare const ParametersSourceTypeEnum: {
+    readonly Text: "text";
+    readonly Youtube: "youtube";
+};
+export type ParametersSourceTypeEnum = typeof ParametersSourceTypeEnum[keyof typeof ParametersSourceTypeEnum];
+export declare const ParametersDetectionModelEnum: {
+    readonly Scrfd500m: "scrfd_500m";
+    readonly Scrfd25g: "scrfd_2.5g";
+    readonly Scrfd10g: "scrfd_10g";
+};
+export type ParametersDetectionModelEnum = typeof ParametersDetectionModelEnum[keyof typeof ParametersDetectionModelEnum];
+export declare const ParametersOutputModeEnum: {
+    readonly PerFace: "per_face";
+    readonly PerImage: "per_image";
+};
+export type ParametersOutputModeEnum = typeof ParametersOutputModeEnum[keyof typeof ParametersOutputModeEnum];
+export declare const ParametersLayoutDetectorEnum: {
+    readonly Pymupdf: "pymupdf";
+    readonly Docling: "docling";
+};
+export type ParametersLayoutDetectorEnum = typeof ParametersLayoutDetectorEnum[keyof typeof ParametersLayoutDetectorEnum];
+export declare const ParametersVlmProviderEnum: {
+    readonly Google: "google";
+    readonly Vllm: "vllm";
+};
+export type ParametersVlmProviderEnum = typeof ParametersVlmProviderEnum[keyof typeof ParametersVlmProviderEnum];
+export declare const ParametersSegmentationMethodEnum: {
+    readonly Scene: "scene";
+    readonly Srt: "srt";
+    readonly Time: "time";
+};
+export type ParametersSegmentationMethodEnum = typeof ParametersSegmentationMethodEnum[keyof typeof ParametersSegmentationMethodEnum];
+export declare const ParametersPdfExtractionModeEnum: {
+    readonly PerPage: "per_page";
+    readonly PerElement: "per_element";
+};
+export type ParametersPdfExtractionModeEnum = typeof ParametersPdfExtractionModeEnum[keyof typeof ParametersPdfExtractionModeEnum];
+export declare const ParametersVisualEmbeddingUseCaseEnum: {
+    readonly Lecture: "lecture";
+    readonly CodeDemo: "code_demo";
+    readonly Tutorial: "tutorial";
+    readonly Presentation: "presentation";
+    readonly Dynamic: "dynamic";
+};
+export type ParametersVisualEmbeddingUseCaseEnum = typeof ParametersVisualEmbeddingUseCaseEnum[keyof typeof ParametersVisualEmbeddingUseCaseEnum];
+//# sourceMappingURL=parameters.d.ts.map