BastionChat

BastionChat

Chat with Documents Privately

0 ratings

Archived App (Last seen on 30 Jul 2026)

This is an archived listing of the app previously available on the App Store.

Although the app is no longer distributed by Apple, you can still view its description, screenshots, version history, ratings, and metadata for reference.
Rating summary

Details

  • Released
  • Updated
  • July 9, 2025
  • June 7, 2026

Features

BastionChat screenshot #1 for iPhone
BastionChat screenshot #2 for iPhone
BastionChat screenshot #3 for iPhone
BastionChat screenshot #4 for iPhone
BastionChat screenshot #5 for iPhone
BastionChat screenshot #6 for iPhone
BastionChat screenshot #7 for iPhone
BastionChat screenshot #8 for iPhone
BastionChat screenshot #9 for iPhone
BastionChat screenshot #10 for iPhone
iphone
ipad
🖼️Get Icon
Icons↘︎

About

Your data belongs to you. Your AI should too. Bastion Chat puts a powerful, private AI assistant in your pocket — no subscriptions, no usage limits, no servers watching over your shoulder. Everything runs locally on your device, so your conversations stay completely confidential. Even offline. ——— KEY FEATURES ——— YOUR PRIVATE AI, YOUR RULES Experience top-tier reasoning and "thinking" models without compromising an ounce of privacy. All processing happens on-device. Your conversations are never uploaded, never analyzed, never sold. Just you and your AI. CURATED MODEL CATALOG Browse and download a hand-picked selection of the best open-source models — from compact 1B models that run on any iPhone, to powerful 27B models on iPad and Mac. Thinking models, vision models, tool-use specialists — all pre-configured and ready to run. VISION & MULTIMODAL AI Ask questions about images, analyze screenshots, describe photos, or extract text from documents. Bastion Chat supports multimodal models that see and understand the world alongside you. AGENTIC AI — YOUR AI THAT ACTS Go beyond chat. Enable Agent Mode to let Bastion Chat reason through complex tasks, plan multi-step solutions, and execute actions autonomously using built-in tools. Your private AI that doesn't just talk — it does. FULL RAG & DOCUMENT INDEXATION A complete knowledge pipeline in the palm of your hand. Import PDFs, web pages, Wikipedia articles, or plain text. Bastion Chat indexes your personal knowledge base using full retrieval-augmented generation so answers are always grounded in your content. AI-POWERED SUMMARIES IN SECONDS Overwhelmed by lengthy articles or dense reports? Bastion Chat distills complex content into clear, actionable summaries. Grasp the key points instantly and save hours of reading. SMART WEB SEARCH, ON DEMAND Need real-time information? Enable the optional web search tool to ground your answers in live results — without sacrificing your privacy or your workflow. BRING YOUR OWN MODEL Import any GGUF model directly from Hugging Face or your local files. Capabilities like advanced reasoning, tool use, and vision are detected and configured automatically on import. SCALES WITH YOUR DEVICE On iPhone, run fast and efficient models for everyday tasks. On iPad and Mac, unleash frontier-class models like Qwen 27B for deep research, long-form writing, and complex reasoning — all still 100% local. ONE PURCHASE. YOURS FOREVER. No usage caps. No API costs. Pay once and run unlimited queries for as long as you own your device.
Show more

What's New in BastionChat

1.1.0

June 7, 2026

This is the biggest update to Bastion Chat yet. VISION & IMAGE UNDERSTANDING Ask questions about photos and images. Multimodal models can now see, describe, and analyze images directly in your conversations. Image history is preserved across sessions. AGENTIC MODE Bastion Chat can now act, not just chat. Enable Agent Mode for multi-step reasoning with autonomous tool use — your AI plans, executes, and delivers results. WEB SEARCH Optionally connect to the web for real-time information. Ground answers in live results without leaving the app or compromising your privacy. WIKIPEDIA IMPORT Pull any Wikipedia article directly into your knowledge base and query it instantly with full RAG. BUILT-IN KNOWLEDGE PACKS Curated, system-managed knowledge packs (first one available: Pet Care) come pre-loaded and searchable — no setup required. SMARTER MODEL BROWSER Redesigned with collapsible sections, family grouping, inline quantization picker, and quality tiers so you always know which model fits your device best. FAVORITES Star your most-used models for instant access at the top of the model list. PER-MODEL SAMPLING PROFILES Fine-tune temperature, context length, and inference settings independently for each model. SMARTER CUSTOM MODEL IMPORT Auto-detects capabilities (reasoning, tool calls, vision) on import. No manual configuration needed. TURBOQUANT KV CACHE New TurboQuant memory modes (3.5-bit and 2-bit) let you run larger models with dramatically less RAM — without sacrificing response quality. FLASH ATTENTION Flash Attention is now enabled by default on all Apple Silicon devices, significantly reducing memory usage during long conversations and improving inference speed. FULL METAL GPU OFFLOAD All model layers now offload to the Metal GPU by default, with device-aware thread and micro-batch tuning for the fastest possible inference on your iPhone, iPad, or Mac. GEMMA 4 THINKING SUPPORT Native support for Gemma 4's thinking block format, including proper stop sequence handling and streamed thinking output. Bug fixes: MCP crash fix, Metal SET_ROWS crash on hybrid GDN models (Qwen 3.6 27B), atomic interrupt flag, download integrity checks, and Qwen context length handling.

More