OpenIntelligence

工具程式

免費 · App 內購買 · 專為 iPad 設計

Turn your documents into an offline neural knowledge base. Stop digging through PDFs. OpenIntelligence brings enterprise-grade RAG (Retrieval-Augmented Generation) to your iPhone—running entirely on your device. 25-STEP RETRIEVAL PIPELINE. ZERO CLOUD DEPENDENCY. This is not simple keyword search. OpenIntelligence orchestrates a rigorous 25-step pipeline used by high-end research systems: • HyDE (Hypothetical Document Embeddings) — Hallucinates an answer to find better semantic matches. • Hybrid Search — Fuses BM25 keywords with 384-dim neural embeddings for maximum recall. • Reciprocal Rank Fusion — Mathematically combines retrieval strategies to surface the best hits. • MMR Diversification — Filters redundant chunks to maximize information gain. • Cross-Encoder Reranking — Scores every candidate with a dedicated neural model for precision. • Graph Context Packing — Optimizes the limited token budget using graph theory. • Parent Document Retrieval — Expands matched sentences to their full paragraph context. • Lost-in-Middle Mitigation — Reorders context to ensure the AI sees the most critical evidence. • Semantic Chunking — Splits documents by logic and structure, not just character counts. • Multi-Column OCR — Metal-accelerated vision processing for complex PDF layouts (360 DPI). • Agentic Tool Use — Autonomous search, summarization, and comparison capabilities. • Motherboard HUD — Real-time X-ray overlay showing Apple Silicon chip positions with live telemetry. • Rich Response Formatting — Full markdown rendering with headers, bullets, bold text, and code blocks. • Device-Optimized Performance — 3-tier Metal GPU shaders, device-specific OCR concurrency, and concurrent neural reranking adapt to your Apple Silicon chip. THE RESULT? Ask "What oil does my car take?" and get "SAE 0W-20 synthetic, 5.3 quarts"—with a direct citation you can tap to verify. DEEP THINK MODE Standard mode answers in seconds. Deep Think spawns multiple reasoning sessions (Self-RAG), cross-validates evidence, and synthesizes a thorough response. Perfect for legal, medical, or academic research. 100% PRIVATE BY DEFAULT Everything runs on-device. Your documents never leave your iPhone unless you enable Private Cloud Compute—Apple's zero-retention fallback. No accounts. No uploads. No tracking. SUPPORTED FORMATS PDF (native + OCR), Word, Excel, PowerPoint, Markdown, CSV, RTF, TXT, and images. SUBSCRIPTION OPTIONS • Pro Monthly: Unlimited documents, all features • Pro Annual: Save 30% compared to monthly • Lifetime: One-time purchase, forever access • Free tier includes core functionality with document limits Payment is charged to your Apple ID account at purchase confirmation. Subscriptions automatically renew unless canceled at least 24 hours before the end of the current period. Manage subscriptions in Settings > Apple ID > Subscriptions. Privacy Policy: https://gunzino.me/openintelligence/privacy Terms of Use: https://gunzino.me/openintelligence/terms

  • 此App未收到足夠的評分或評論以顯示概況。

Version 1.2.0 RICH MARKDOWN RESPONSE RENDERING Responses now display with full formatting — headers, bullets, numbered lists, bold, code blocks, and block quotes. • Block-Level Markdown Parser — h1-h6, bullet lists, numbered lists, code fences, block quotes, horizontal rules, paragraphs. • Inline Normalization — Apple FM concatenates markdown on one line. 6 regex patterns split headers, bullets, numbered items onto separate lines before parsing. • Response-Cleaning Functions Audited — Two functions were stripping ALL markdown. Both rewritten to preserve formatting while removing artifacts. • Formatting-Aware LLM Prompts — All system prompts now instruct the model to use headers, bullets, and bold. • Paragraph Structure Preservation — Sentence joining uses paragraph breaks, maintaining document structure. DEVICE-OPTIMIZED PERFORMANCE ENGINE Every pipeline stage adapts to your specific Apple Silicon chip. • 3-Tier Metal GPU Search — Automatically selects fastest shader: Threadgroup (≥1000 vectors), SIMD4 (100-999), or Scalar fallback. Previously all searches used the slowest kernel. • Device-Specific OCR Concurrency — A19 Pro: 8 ops @ 1ms. A18 Pro: 6 @ 2ms. A17 Pro: 4 @ 3ms. Older: 2 @ 6ms. • Adaptive OCR Page Filtering — Pre-screens pages, skipping 50-80% of clean digital pages before OCR runs. • Concurrent GPU Rendering — CIFilter preprocessing now runs in parallel (was serialized). • Concurrent Neural Reranking — Cross-encoder scores pairs via TaskGroup. Bulk dataPointer copy instead of per-element subscript — 3× faster. • GPU Embedding Ingestion Mode — Uses CPU+GPU compute units during ingestion, freeing Neural Engine for OCR. • 5-Candidate OCR — Evaluates 5 transcription alternatives per recognition (was 3). STABILITY & HARDENING • Fixed crash in MMR diversification (dimension 0 edge case) • Matrix dimension validation added to RAGEngine • GPUComputeService edge cases — empty arrays no longer wrapped in spurious outer arrays • StoreKit product loading races against 5-second timeout — no more 30-60s hangs on airplane mode • Fatal threading assertion replaced with safe fallback in Release builds • Pipeline trace diagnostics compiled out of Release via #if DEBUG • Privacy manifest updated with SystemBootTime API declaration • Dead code removed from cross-encoder path ZERO-DATA-LOSS INGESTION Prevents silent content loss on font-encoded PDFs. • Font Cipher Detection — PHASE -1 validation detects PDFs where characters are shifted (Kia, Hyundai manuals). Jaccard similarity between OCR and PDFKit text. Garbled docs route through full Vision OCR. Prevents 93% content loss. • Raw String Regex Fix — 5 Unicode patterns were failing in Swift raw strings. CJK bullet, en-dash, em-dash patterns now work. • Garbled Image Extraction Fix — Image extraction on font-encoded PDFs no longer skips diagrams. • Dynamic Image Text Budget — Scales with document complexity instead of fixed limit. SWIFT 6 COMPLIANCE • 11 files updated with strict concurrency annotations. Zero runtime behavior change.

開發者Gunnar Hostetler表明App的私隱慣例或包括下列資料的處理。詳情請參閲 開發者的私隱政策

  • 不收集資料

    開發者不會從此 App 收集任何資料。

    私隱慣例或因使用的功能或年齡等因素而異。 進一步了解

    開發者尚未表示此 App 支援哪些輔助使用功能。 進一步了解

    • 供應商
      • Gunnar Hostetler
    • 大小
      • 84 MB
    • 類別
      • 工具程式
    • 相容性
      須使用 iOS 26.0 或以上版本。
      • iPhone
        須使用 iOS 26.0 或以上版本。
      • iPad
        須使用 iPadOS 26.0 或以上版本。
      • Mac
        須使用 macOS 26.0 或以上版本及配備 Apple M1 或以上版本晶片的 Mac。
    • 語言
      • 英文
    • 年齡分級
      16+
    • App 內購買
      • Pro Monthly USD 5.99
      • Document Pack USD 2.99
      • Pro Annual USD 49.99
      • Lifetime Cohort USD 59.99
    • 版權
      • © 2026 Gunnar Hostetler