oriveo/oriveo

Open-source, bring-your-own-key (BYOK) multi-model AI chat client for iOS, Android and the web. One LLM client for OpenAI, Anthropic, Google Gemini, OpenRouter, DeepSeek and ten more providers, plus any OpenAI-, Anthropic- or Gemini-compatible endpoint — Ollama, LM Studio, llama.cpp, vLLM. Local-first, self-hostable, no account. AGPL-3.0.

What it solves

Oriveo Community Edition is a local-first, open-source AI chat client that allows users to interact with multiple AI models from various providers using their own API keys (BYOK). It eliminates the need for a middleman subscription, ensuring users pay model providers directly at list price without markups or tracking.

How it works

The project provides three native clients (iOS, Android, and Web) that share a common request specification. Users provide their own API keys for 15 built-in providers or connect to local model servers (like Ollama or vLLM) via compatible endpoints.

  • iOS and Android: Requests go directly from the device to the provider.
  • Web: Requests typically pass through a self-hosted Next.js route handler to bypass CORS restrictions, though some endpoints are called directly.
  • Local Storage: Conversations, notes, and attachments are stored on the device rather than in a cloud account.

Who it’s for

It is designed for users who prefer a private, account-free experience and want to a single interface to manage multiple AI models and providers without paying for a third-party subscription service.

Highlights

  • Multi-Provider Support: Native integration with 15 providers including OpenAI, Anthropic, Google Gemini, and DeepSeek.
  • Local-First Privacy: No accounts required; data stays on the device with manual export/backup options.
  • Local Model Integration: Supports llama.cpp, Ollama, LM Studio, and vLLM.
  • Advanced Chat Features: Includes reasoning blocks, citations, attachment support (PDF, Office, EPUB, etc.), and MCP tool integration.
  • Cross-Check: Ability to hand an answer to a second model for review.
  • Memory: Persistent facts about the user carried into new conversations.
  • On-Device Cost Tracking: Computes spend per message and provider based on actual response reports.

Related

  • Project
    haowjy/creative-writing-skills

    An AI‑driven creative‑writing assistant built as Claude‑compatible agents and skills. Muse coordinates brainstorming, drafting, critique, editing, and a growing knowledge base, letting writers maintain voice and continuity across a story.

  • Project
    microsoft/LangChain4j-for-Beginners

    LangChain4j for Beginners is a free, multi‑language course that teaches Java developers how to build AI applications with LangChain4j and Azure OpenAI, progressing from basic chat to advanced RAG and AI agents. The repository includes code examples, video walkthroughs, and a pre‑configured Codespace environment.

  • Project
    tonykipkemboi/ollama_pdf_rag

    A local RAG application that lets users chat with multiple PDF documents using Ollama and LangChain, featuring Next.js and Streamlit interfaces.

  • Project
    dappros/ethora

    Ethora is a multi‑platform SDK suite that adds real‑time chat, push notifications, file sharing, and LLM‑powered AI bots to web, mobile, and WordPress apps. It ships as a monorepo of sub‑modules (React, React‑Native, Android, iOS, WordPress) and can run on Ethora’s managed cloud or be self‑hosted via Docker/Helm. The platform includes a REST + XMPP backend, MCP server for AI‑agent integration, RAG demos, and optional blockchain wallet features. Getting started is as simple as cloning the repo, running `npx @ethora/setup` to obtain credentials, and installing the SDK for your target platform.