Spike AI vs Bring Your Own AI

Meet Spike Studio: AI chat on your iPhone, with a choice of where your model runs. GET STARTED WITH SPIKE CLOUD Create a free account in the app, verify your email, and start chatting. You do not need a computer or your own server. Free usage limits apply. This version offers no in-app subscriptions or purchases. USE YOUR OWN MODELS Connect an OpenAI-compatible server, including LM Studio, or download an on-device model on a supported iPhone. These options do not require a Spike account. Model downloads need available storage, and on-device performance depends on your iPhone and the selected model. CONNECT YOUR COMPUTER Use an account-linked Runtime from Spike Studio on your computer, and follow Agent work from your iPhone. Computer and Runtime setup are required for these optional features. KEEP YOUR CONVERSATIONS TOGETHER Sign in with the same Spike account across devices to sync encrypted conversations. Organize chats in projects and choose tools for the task at hand. Tools that use iPhone features request the relevant permissions. Cloud chats are processed by Spike Cloud; on-device models run on your iPhone; a connected server processes the chats you send to it. Choose the option that fits your needs.

Bring Your Own AI is a private, bring-your-own-key chat client for the world's most capable AI models. Connect OpenAI, Anthropic, DeepSeek, Groq, OpenRouter, Ollama, or any OpenAI-compatible endpoint — plus Apple's on-device Foundation Model — and chat with all of them from a single, beautifully native iOS interface. WHY BRING YOUR OWN AI • BYOK (Bring Your Own Key). Your API keys live in the iOS Keychain, encrypted with Secure Enclave-backed AES-256. They never touch our servers. There are no servers. • Every model, one app. Switch between GPT-4o, Claude 3.5 Sonnet, DeepSeek-V3, Llama 3, Qwen, your local Ollama instance, and Apple Intelligence without juggling browser tabs. • Thinking, visible. Reasoning models (DeepSeek-R1, o1, etc.) expose their chain-of-thought — collapse or expand on demand. • Streaming. Watch responses render token by token, with full Markdown, LaTeX, and syntax-highlighted code blocks. • Auto-title. Conversations name themselves from the first exchange — never wade through "New Chat #47" again. WHAT YOU CAN DO • Multi-provider endpoint manager — test connectivity, edit base URLs, rotate keys, all in one place. • Per-conversation model selection — use DeepSeek for code, Claude for writing, GPT-4o for vision, all in the same session. • Code blocks with syntax highlighting (Swift, Python, JS, Go, Rust, 100+ languages). • Markdown rendering for headings, lists, tables, links, and inline math. • Image attachments to vision-capable models (GPT-4o, Claude 3.5 Sonnet, etc.). • Export any conversation as Markdown or JSON. • Siri shortcuts: "Hey Siri, new Bring Your Own AI conversation." • Local-first. Every conversation lives only on your device. AirDrop to back up. PRIVACY Bring Your Own AI does not collect any data. We have no analytics SDK, no telemetry, no remote logging. All conversations are stored locally in an encrypted SQLite database. All API calls go directly from your device to your chosen provider — we never see your prompts or responses. The app is open-source. REQUIREMENTS • iOS 26.0 or later • iPhone only • Apple Intelligence features require iPhone 15 Pro or later (A17 Pro / A18+) • A valid API key from at least one supported provider Bring Your Own AI is built by people who believe AI should be a tool, not a walled garden.