TypingMind is a chat app for AI models where you bring your own API keys and pay the AI providers directly for what you use. The app itself is sold as a one-time personal license or as a monthly team workspace, and no plan includes AI usage. What decides whether it suits you is whether you are comfortable managing API keys and per-token bills in exchange for one interface across many providers.
What is TypingMind?
TypingMind is run by Typing Technology Co. Ltd. Its founder, Tony Dinh, started it as an advanced chat interface, and it is now a front end for models from many providers. It runs in the browser and installs as a web app on Mac, Windows, Linux, iOS and Android.
The personal app keeps your data on your device and needs no account, with cloud sync as an option. After you buy a license you also get an offline static web app to host yourself. TypingMind Teams is a hosted workspace, and it can be self-hosted too.
What you can do with TypingMind
TypingMind supports models from 12 AI providers: OpenAI's GPT models, Anthropic's Claude models, Google's Gemini models, DeepSeek, Mistral, Moonshot Kimi, Perplexity Sonar, xAI Grok, Z.ai GLM, OpenRouter, Groq and local models through Ollama or LM Studio. You add an API key for each provider, the model list follows each provider's catalog, and custom endpoints such as Azure, AWS Bedrock and Hugging Face can be added. TypingMind has no models of its own.
Web search comes on the Extended license and up, through search plugins such as Perplexity Search. A Deep Research plugin plans multi-step research with live web search and content analysis, and it needs your own SerpAPI and Firecrawl API keys. Extended and up also add uploads of PDF, DOCX, XLSX and Markdown documents, vision and image generation through plugins such as GPT Image.
The built-in Toolbox plugin has the AI write Python or JavaScript, run it in your browser on your device and give you the result in the chat. Memory lets TypingMind remember useful information about you across conversations. Project Folders, on the Premium license and up, build dedicated workspaces with their own chat history and knowledge bases.
You can dictate messages with your microphone, with optional auto-send, and have replies read aloud, but there is no live spoken conversation mode. Plugins and MCP servers connect TypingMind to Slack, Zapier, GitHub, Notion, Google Drive, Figma, Make, n8n, Todoist, Supabase, Firecrawl and Azure AI Search, and the knowledge base can pull from GitHub, Google Drive and Notion.
Working with other people
TypingMind Teams is an AI workspace for your team, sold apart from the personal licenses. Teams plans add an Integration API, and the Teams Professional plan and up add single sign-on, directory sync and OAuth 2.
TypingMind pricing: what each plan gives
No TypingMind plan includes AI usage. Model costs are billed per token by OpenAI, Anthropic, Google or another provider to your own API key.
The Free version works with your own OpenAI API key, but chat history is not saved, a popup appears every few minutes and some other features are limited.
Personal licenses are one-time purchases with no renewal, and each license key works on 5 devices. The Standard license costs $39, the Extended license $79 and adds web search, file uploads and image generation, and the Premium license $198 and adds Project Folders. A Bulk license for 10 users costs $790. Personal licenses come with a 14-day money-back guarantee.
TypingMind Teams plans include 5 seats each, with extra seats at $8 per seat a month. Teams Starter costs $99 a month, or $83 a month billed yearly; Teams Growth costs $199 a month, or $166 billed yearly; and Teams Professional costs $599 a month, or $499 billed yearly. Teams plans start with a 14-day free trial. Business is priced on request, from $18,700 a year.
Privacy and security
TypingMind does not train on your chats: "All of your conversations on the TypingMind system are strictly confidential and never used for training purposes." What happens to your prompts at the AI provider depends on that provider's API terms.
TypingMind is SOC 2 Type 2 certified and has a GDPR statement and a Data Processing Agreement. Single sign-on is on Teams Professional and up. Support runs by email at support@typingmind.com and through a help center.
Who TypingMind suits
TypingMind suits people who already have API keys, or are happy to get them, and want one chat interface across many providers, including local models, while paying only for the tokens they use. Its one-time personal licenses, on-device data and self-hosted option suit anyone who wants to buy the app once and keep it.
It suits less well anyone who wants a single bill that covers AI usage, since every plan leaves the model costs to your own API keys. Web search, file uploads and image generation need the Extended license or higher, Deep Research needs further API keys, and the team plans are a separate monthly product starting at 5 seats.
Frequently Asked Questions
How much does TypingMind cost?
Personal licenses are one-time payments: $39 for Standard, $79 for Extended and $198 for Premium. Teams plans start at $99 a month for 5 seats, or $83 a month billed yearly. AI usage is paid separately to the provider through your own API key.
Which AI models can I use in TypingMind?
Models from OpenAI, Anthropic, Google, DeepSeek, Mistral, Moonshot, Perplexity, xAI, Z.ai, OpenRouter and Groq, plus local models through Ollama or LM Studio and custom endpoints such as Azure, AWS Bedrock and Hugging Face. Which models appear depends on the API keys you add.
Is TypingMind a one-time purchase?
The personal licenses are: you buy once with no renewal, and each key works on 5 devices. TypingMind Teams is a monthly or yearly subscription.
Does TypingMind train AI on my chats?
No. Conversations on TypingMind are never used for training, and what the AI provider does with your prompts depends on its own API terms.
