Now in public beta

One prompt.
Every model.
Think deeper.

The AI chat platform for power users. Connect Claude, GPT, Gemini, Grok, Mistral, and more — all in one place. Bring your own API keys. Pay only for what you use.

Get Started FreeSee Features
go.perpendo.app
Claude 4 SonnetGPT-4oGemini 2.5 Pro
Claude 4 Sonnet
GPT-4o
Gemini 2.5 Pro

Multi-model AI chat — built for people who think for a living

AnthropicOpenAIGoogleMistralGroqxAI

What is Perpendo?

Perpendo is a BYOK AI chat platform that lets you use Claude and GPT in one interface — along with Gemini, Grok, Mistral, and any OpenAI-compatible endpoint. Bring your own API keys, pay only for what you use, and keep full control of your data.

If you’re looking for a ChatGPT alternative with your own API key or a TypingMind alternative with more features, Perpendo offers conversation branching, side-by-side model comparison, AI prompt chains, a command palette, and a full analytics dashboard — all cloud-hosted with nothing to self-host.

What is the best way to use multiple AI models? With Perpendo you can compare AI model responses by sending one prompt to 2–4 models simultaneously and watching them stream in parallel. Switch models mid-conversation, fork branches to explore different directions, and track costs across every provider from a single dashboard.

Perpendo is a privacy-first, encrypted AI chat platform where your API keys are protected with AES-256-GCM and your conversations are never read, sold, or used for training. The self-hosted license adds full Ollama and local model support — run inference on your own hardware with zero API costs.

What makes Perpendo different

Every feature power users wish they had. All in one place.

Multi-Model Chat

Switch between Claude, GPT, Gemini, Grok, and more mid-conversation. Each response shows which model generated it.

Side-by-Side Comparison

Compare your chosen models with your own API keys. Responses are saved, branchable, and private — not anonymous benchmarks.

Conversation Branching

Fork any conversation at any point. Explore different directions. Navigate branches with a visual tree.

Command Palette

Cmd+K to find anything — conversations, prompts, agents, settings. Fuzzy search across everything you’ve ever discussed.

AI Agents

Save your favorite system prompt + model + parameter combos as one-click personas. Research Assistant, Code Reviewer, Writing Editor — switch instantly.

Prompt Library

Build reusable prompts with {{variables}} that present input forms when triggered. Share across conversations.

Web Search

Toggle search-enhanced answers. Perpendo searches the web, injects results as context, and the AI cites its sources.

Prompt Chains

Multi-step workflows where each step’s output feeds the next. Research → Outline → Draft in one click.

Analytics Dashboard

Track token usage, costs, and response times across all models. Export as CSV. Know exactly what you’re spending.

Voice Input

Dictate messages with your voice. Perfect for mobile use.

Telegram Bot

Chat with your AI models from Telegram. Send voice messages, images, switch models — all from your phone.

Encrypted by Default

API keys encrypted with AES-256-GCM. Your conversations are yours. We never see your API keys or train on your data.

Perpendo vs the rest

See how Perpendo stacks up against the alternatives.

FeaturePerpendoSelf-HostedTypingMindChatGPTClaude
Multi-model supportAll major providersAll major providersAll major providersOpenAI onlyAnthropic only
Bring your own keys
Conversation branchingFull tree navigatorFull tree navigatorBasic
Side-by-side comparison2–4 models parallel2–4 models parallel
Command paletteYes (Cmd+K)Yes (Cmd+K)
AI Agents / PersonasCharactersCustom GPTsProjects
Prompt chains
Web search integrationYes (Tavily)Yes (Tavily)Plugin-basedBuilt-inBuilt-in
Telegram bot
Voice input
Local model supportSelf-hosted planFull (Ollama)
API key encryptionAES-256-GCMAES-256-GCMlocalStorageN/AN/A
Analytics / cost trackingFull dashboardFull dashboardBasic
DatabasePostgreSQLSQLite (zero config)BrowserProprietaryProprietary
Keyboard shortcutsComprehensiveComprehensiveLimitedLimitedLimited
PriceFrom $0/mo (BYOK)$49 one-time$23–47 one-time$20/mo fixed$20/mo fixed

How it works

Up and running in under a minute.

01

Sign up free

Create your account in seconds. No credit card required.

02

Add your API keys

Connect Claude, GPT, Gemini, or any model. You only pay providers for what you use.

03

Start thinking deeper

Chat, compare, branch, and build. Your AI workspace, your way.

Simple, transparent pricing

Start free. Upgrade when you need more.

MonthlyYearlySave 35%

Free

$0/month
  • 2 AI providers
  • 10 conversations
  • Basic chat with streaming
  • Markdown rendering
  • Keyboard shortcuts
Get Started Free
Most Popular

Plus

$39/year

$3.25/mo billed annually

  • Everything in Free, plus:
  • All AI providers (7+)
  • Unlimited conversations
  • Conversation branching
  • Side-by-side comparison
  • AI Agents & Personas
  • Prompt library with variables
  • Analytics dashboard
  • Canvas / artifacts editor
  • Voice input
  • Conversation tags & archive
  • Theme customization
Start Plus

Pro

$79/year

$6.58/mo billed annually

  • Everything in Plus, plus:
  • Telegram bot integration
  • Voice transcription
  • Prompt chains
  • Web search integration
  • Anthropic prompt caching
  • Priority support
Start Pro
One-Time Purchase

Self-Hosted

$49one-time
  • All Pro features
  • Run on your own hardware
  • Full Ollama / local model support
  • SQLite database (zero config)
  • Lifetime updates
  • No recurring fees
  • Up to 3 devices
Buy License

Perfect for developers and privacy-first users who want full control.

Frequently asked questions

What does "bring your own keys" mean?

You connect your own API keys from providers like Anthropic, OpenAI, Google, etc. You pay them directly for usage. Perpendo is the interface — we never charge you for AI usage.

Is my data private?

Yes. Your API keys are encrypted with AES-256-GCM. Your conversations are stored in an isolated database. We never see your keys, read your conversations, or train on your data.

How is this different from ChatGPT or Claude?

ChatGPT and Claude lock you into one provider at $20/month. Perpendo lets you use every major AI model from one interface, pay only for what you use (often much less than $20/month), and switch between models mid-conversation.

How is this different from TypingMind?

Perpendo is cloud-hosted (no self-hosting required), supports conversation branching, side-by-side model comparison, prompt chains, a Telegram bot, and full analytics — features TypingMind doesn’t offer.

Can I use local models?

Yes. Purchase the self-hosted license ($49 one-time) and run Perpendo on your own machine. Connect Ollama or any OpenAI-compatible endpoint via the Custom provider. Run models like Llama 3, Gemma, and Phi-4 on your own hardware with zero API costs and full privacy. The cloud version at go.perpendo.app connects to remote AI providers via API keys.

Is there a free tier?

Yes. The free plan includes 2 providers and 10 conversations. No credit card required.

What’s the difference between cloud and self-hosted?

The cloud plans (Free, Plus, Pro) run at go.perpendo.app — we handle hosting, updates, and backups. The Self-Hosted plan is a one-time $49 purchase that lets you run Perpendo on your own hardware with a SQLite database, full Ollama / local model support, and no recurring fees. Both options include all Pro features.

Can I cancel anytime?

Yes. No contracts. Cancel from your billing dashboard and your plan reverts to free at the end of the billing period.

Ready to think deeper?

Join Perpendo and take control of your AI workflow.

Get Started Free

No credit card required. Free plan available.