Workspace for artifacts and files
Create presentations, documents, code, and other files with Canvas, Python, npm packages, and browser tools available alongside the thread.
ai.diy
ai.diy is a browser-owned, open-source AI workspace for BYOK chat and local-first work. It keeps chats, files, knowledge, and settings in your browser while relaying requests to 20+ cloud and local providers through a self-hosted architecture.
Overview
ai.diy is a browser-owned, open-source AI workspace built for bring-your-own-key chat and local-first work. It keeps workspace state in the browser while routing requests through a self-hosted relay, so chats, files, knowledge, and settings remain under the user’s control rather than living in a provider-managed workspace.
The product is designed for people who want to work across cloud and local models without rebuilding context every time they switch tools. It supports 20+ providers and includes Canvas, Python, npm packages, browser tools, and local knowledge features for tasks such as writing documents, creating presentations, reviewing code, and assembling useful artifacts.
Core features
Browser-owned persistence
Keep chats, files, knowledge, and settings in the browser so workspace state persists locally instead of in a provider account.
Multi-provider BYOK chat
Connect to 20+ cloud and local models, then switch providers mid-thread without moving the workspace or losing context.
Self-hosted relay
Use a relay architecture where provider credentials stay in browser storage and are forwarded per request rather than kept as persistent server secrets.
Local RAG for private documents
Search notes and PDFs with on-device embeddings; the index stays local, while retrieved context can be sent to the chosen model.
Integrated tools and agents
Use DuckDuckGo, Firecrawl, Parallel MCP, browser Python, remote MCP, slash skills, approved subagents, and website presets for research and agent workflows.
Typical use cases
Create working artifacts
Draft presentations, documents, code snippets, and other files in one place while keeping thread context, memory, and workspace state available in the browser.
Move between AI providers
Switch between cloud and local providers during a conversation when you want to compare model behavior or use a specific model for a task without starting over.
Work with local knowledge bases
Search notes and PDFs with local embeddings for private document retrieval, then send only the retrieved context to the selected model when needed.
Run a self-hosted AI workspace
Use the workspace as a self-hosted relay-backed environment when you want control over deployment rather than relying on a single vendor-managed AI app.
Handle coding and review tasks
Review code or refactor diffs with BYOK access or local models such as Ollama, keeping the work inside the browser workspace rather than a vendor-specific editor.
Pros and Cons
Pros
- Workspace state stays in the browser, including threads, Canvas, memory, knowledge, and files.
- Supports 20+ cloud and local providers, making it easier to switch models without moving work between apps.
- Open-source and MIT licensed, with self-hosting options via Node.js, Docker Compose, or Vercel preview.
- No persistent server-side LLM credentials are required; provider keys remain in browser storage and are relayed per request.
- Includes built-in tools for artifacts, research, local document search, and agent workflows.
Cons
- The site notes that a hosted operator can observe traffic in transit, so users who need a tighter network boundary should self-host.
- The source does not show public pricing or plan details, and the /pricing route returns a 404 page.
FAQ
What is ai.diy?
ai.diy is an open-source, self-hosted AI workspace for chat, research, tools, Canvas artifacts, memory, and local knowledge. It uses bring-your-own-key access so you can choose the provider and model rather than using a single hosted AI service.
Are my provider keys stored on the ai.diy server?
No server-side LLM credentials are required. Provider keys are kept in your browser and relayed per request to the provider you select. The site notes that a hosted operator can still observe traffic in transit, so self-hosting is the option for controlling the infrastructure and network boundary.
Where does ai.diy store my workspace data?
Chats, files, Canvas artifacts, memory, on-device knowledge-base chunks, usage events, and preview sessions persist in your browser through IndexedDB and localStorage. Optional backups to S3, WebDAV, or Google Drive are client-side features you can enable yourself.
Which AI providers work with ai.diy?
ai.diy supports more than 20 integrations, including OpenAI, ChatGPT subscription, Anthropic, Google Gemini, Groq, Cerebras, Fireworks, Perplexity, Cohere, OpenRouter, DeepSeek, xAI, Ollama, Mistral, Hugging Face, Amazon Bedrock, Azure, Vertex, Vercel Gateway, Together, LM Studio, and custom OpenAI-compatible endpoints.
Can I self-host ai.diy?
Yes. You can run the production build on a standard Node.js server or use Docker Compose. The server acts as a request relay and does not need provider API keys in environment variables.
Quick Facts
- Category
- AI workspace
- Platform
- Browser-based web app
- Ownership model
- Local-first, browser-owned state
- Licensing
- MIT licensed, open source
- Hosting
- Self-hostable on Node.js, Docker Compose, or Vercel preview
- Source domain
- tryaidiy.com
Alternativas ao ai.diy
Goldfish
Goldfish is an AI memory app for macOS that helps you reply, write, summarize, and continue using the context already on your Mac. It stores memory locally and writes in your tone.
ArtDeck
ArtDeck is a visual reference board app for iPhone, iPad, and Mac that lets artists collect mixed reference material on one canvas and study it with ToolBox lenses. It works offline, supports optional iCloud sync, and is sold as a one-time purchase with no account required.
Claude Overlay
Claude Overlay is a Windows desktop overlay for Claude Code that reads your screen so you can ask questions, inspect content, and request edits without leaving the current app. It runs on your existing Claude subscription through the Claude CLI, so no separate API key is required.
Dcyde
Dcyde is a decision log for product teams that captures what was decided, the reasoning behind it, and the source in one shared place. It supports pinning decisions from Slack, Figma, or inside Dcyde and keeps the record searchable over time.
Lasso
Lasso is an ecommerce product data platform for enriching catalog records, processing supplier files, generating product content, and monitoring competitors. It combines a web app with a REST API, SDK, and MCP server for teams and developers.
Biji
Biji é uma plataforma versátil projetada para aumentar a produtividade por meio de ferramentas e recursos inovadores.