githtml vs On Device

Your AI docs, finally readable. githtml is the iOS reader for HTML documentation that lives inside your GitHub repos. Documentation your AI coding agent writes, the rendered output, the framework references you keep meaning to read on the train. https://www.apple.com/legal/internet-services/itunes/dev/stdeula/ Connect your GitHub, pick a repo, tap a file. That's it. — WHAT IT DOES • A clean WebKit reader that gets out of the way of your content — the doc's own typography and styling stay intact. • A bottom drawer for capturing notes on any page. Auto-saved per file, exportable as plain text or through the system share sheet. • Browse, Favorites, and Recents — all three grouped by owner so a hundred repos stay organized. • Scroll-aware navigation chrome that fades while you read and comes back the instant you scroll up. • Offline cache — once you've opened a doc, the bytes stay on your device. — WHO IT'S FOR • Developers using Claude Code, Cursor, Codex, or other coding agents that write extensive docs into their repos. • Solo builders maintaining a personal knowledge base in private repos. • Engineers reading framework docs, API references, and internal wikis on the train, on planes, or away from Wi-Fi. • Anyone who has tried to read raw Markdown in the GitHub mobile app and given up. — HOW IT WORKS 1. Sign in with GitHub. The app uses GitHub's official App flow — you pick which repos to grant access to. 2. The permission requested is "Read access to code." Nothing else. No write access, no organization data, no actions. 3. Pick an owner, then a repo, then a file. HTML files are indexed by default. — LOCAL-ONLY, ALWAYS • Your GitHub token lives in the iOS Keychain. • Your notes live in local SwiftData storage. • Your cached docs live in the local file system. • "Delete all data" wipes every byte the app has ever stored, in one tap. • There is no githtml server. The only network calls are to GitHub itself and a tiny token-exchange function that never sees your token after the exchange completes. — WHAT'S NEXT githtml is built around a tight V1: read your repos, take notes, work offline. Future versions will add more ways to navigate and search a large archive of cached docs. Email feedback to shape what ships next.

On Device LLM is a local-first AI workbench for your iPhone. Open-source language and vision models run entirely on-device via Apple's MLX framework — no servers, no API keys, no telemetry, no accounts. LIVE CAMERA CAPTION Point your camera at anything. SmolVLM describes the scene every few seconds, streaming the caption right onto the viewfinder. Tune the refresh interval from 500ms to 30s. Tap the refresh icon for an on-demand caption between cycles. VOICE CONVERSATION Hands-free chat with an on-device LLM. Industry-standard voice activity detection ignores TV chatter and background noise; speaks back through the same audio session it's listening on — no clipped first words, no delayed replies. CODE ASSISTANT Pick from Qwen 2.5 Coder, Llama 3.2, Phi-3.5, Gemma 2, Qwen3 and dozens of MLX-converted models on HuggingFace. The app auto-selects a model that fits your device's RAM (1.5B on older iPhones, up to 7B on Pro Max). Streamed token-by-token output, code-block syntax highlighting, conversation export to Markdown. POWER TOOLS • Document scanner — capture multi-page code via the native four-corner doc cam • Live OCR — text on a screen, sign, or whiteboard recognized in real time • A/B compare — run two models side-by-side and watch the speed/quality tradeoff • Benchmark — measure tokens per second, time to first token, peak memory, thermal impact • Macros — chain prompts together (lint → refactor → test) • Snippets — save reusable prompt templates • Persona memory — different facts remembered per persona • Mac Bridge — pair your Mac, use the iPhone as an MLX inference server PRIVACY YOU CAN VERIFY • Every model runs on the device — your prompts never leave the phone • A network-activity indicator surfaces every outbound request (only model downloads) • Conversations stored with at-rest encryption • One-tap "wipe all on-device data" in Settings • No analytics SDKs, no sign-up, no account required OPTIMIZED FOR YOUR DEVICE • First-launch device-tier auto-pick for both the chat model and the camera VLM • SmolVLM 2.2B (bf16) on Pro Max; SmolVLM 4-bit on entry-tier iPhones • Automatic fallback chain when a model mirror is unreachable • Thermal awareness — clamps response length only at .critical (matches peer on-device apps) • Cancels in-flight Metal work when the app backgrounds so iOS won't kill it REQUIREMENTS • iOS 18.0 or later • Recommended: iPhone 12 or newer • Wi-Fi recommended for the initial model download (≈1.5 GB minimum) Open source models. Open standards. None of your data on someone else's server.