blah.chat

Personal AI chat assistant with access to all models (OpenAI, Gemini, Claude, xAI, Perplexity, and more), mid-chat model switching, conversation branching, and transparent cost tracking.

✨ Features

• All Models in One Place: Access GPT-5, Claude Opus 4.5, Gemini 2.5 Pro, Grok, and 50+ models via Vercel AI Gateway
• Mid-Chat Model Switching: Compare responses or switch models without losing context
• Conversation Branching: Fork conversations to explore different directions
• Resilient Generation: Responses survive page refresh, tab close, even browser crashes
• RAG Memory: Automatic semantic memory extraction and retrieval
• Voice Input: Audio transcription with Whisper (OpenAI/Groq)
• AI Tools: Web search (Tavily), code execution (E2B), URL parsing (Jina)
• Cost Tracking: Per-message token usage and cost breakdown
• Real-Time Collaboration: Multi-user projects with live sync (via Convex)

💻 CLI

Chat with AI models directly from your terminal.

Install

# Shell (macOS/Linux)
curl -fsSL https://blah.chat/install | bash

# npm
npm i -g @blah-chat/cli

# Homebrew (coming soon)
brew install planetaryescape/tap/blah

Usage

# First time: authenticate via browser
blah login

# Start chatting
blah "What is the capital of France?"
blah --model claude-3-opus "Explain quantum computing"
blah --help

📸 Preview

<table> <tr> <td width="50%"> <img src="./public/assets/app-preview.jpeg" alt="Dark mode" /> <p align="center"><em>Dark mode</em></p> </td> <td width="50%"> <img src="./public/assets/app-preview-light.jpeg" alt="Light mode" /> <p align="center"><em>Light mode</em></p> </td> </tr> </table>

🔑 Required API Keys

blah.chat requires API keys for certain features:

Core Features (Required)

• AI_GATEWAY_API_KEY - Vercel AI Gateway for all AI model access

Speech Features (Optional)

Speech-to-Text (STT)

Requires ONE of the following providers (configured by admin in Settings):

• GROQ_API_KEY - Groq Whisper Turbo (default, $0.04/hour)
• OPENAI_API_KEY - OpenAI Whisper ($0.006/min)
• DEEPGRAM_API_KEY - Deepgram Nova-3 ($0.0077/min)
• ASSEMBLYAI_API_KEY - AssemblyAI ($0.0025/min)

Text-to-Speech (TTS)

• DEEPGRAM_API_KEY - Deepgram Aura voices (required)

Note: If STT/TTS API keys are not configured, these features will be automatically disabled. Users will see an error message when attempting to enable them:

• Development: Specific missing key name shown
• Production: "Please contact your administrator" message

See the Self-Hosting Guide for full environment variable setup.

🚀 Quick Deploy

Vercel (Recommended - 10 minutes)

Auto-configures Convex and Clerk integrations. Just add your AI_GATEWAY_API_KEY.

Railway (10 minutes)

Railway auto-detects environment variables and guides you through setup.

Self-Hosting

For full instructions on self-hosting (including Fly.io, custom VPS, Docker), see SELF_HOSTING.md.

📄 License

blah.chat is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0).

Self-hosted usage: Free under AGPL-3.0 with default limits (50 messages/day, $10/month budget per user). Commercial usage (exceeding limits or without source disclosure): Requires commercial license. Cloud version: Coming soon with subscription plans.

For commercial licensing or questions, contact: blah.chat@bhekani.com

Why AGPL?

We chose AGPL to:

• Protect the community from exploitation by cloud providers
• Ensure improvements are shared back with the community
• Allow free self-hosting while building a sustainable business

If you modify blah.chat and run it as a network service, AGPL Section 13 requires you to provide users with access to your modified source code.

---

Development Setup

1. Environment Variables

Copy the example environment file:

cp .env.local.example .env.local

You will need to configure the following API keys in .env.local:

Core Services

• Vercel AI Gateway (AI_GATEWAY_API_KEY): Required for all AI model inference.
• Clerk (NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY, CLERK_SECRET_KEY, CLERK_ISSUER_DOMAIN): For user authentication and Convex integration.
• Convex (NEXT_PUBLIC_CONVEX_URL, CONVEX_DEPLOYMENT): Backend database and functions (configured automatically via bunx convex dev).

AI Tools & Integrations

• Tavily (TAVILY_API_KEY): Enables real-time web search capabilities.
• Jina (JINA_API_KEY): Used by the URL Reader tool to parse web pages into markdown.
• E2B (E2B_API_KEY): Powers the code interpreter sandbox for executing code safely.
• Firecrawl (FIRECRAWL_API_KEY): Optional alternative for advanced web scraping/crawling.
• OpenAI (OPENAI_API_KEY): Needed for audio transcription (Whisper) if not using Groq.

2. Convex Environment Variables

The backup backend (Convex) requires these variables to be configured in the Convex Dashboard:

• AI_GATEWAY_API_KEY: Vercel AI Gateway key for model access within Convex actions.
• CLERK_ISSUER_DOMAIN: Your Clerk Frontend API URL (e.g., your-app.clerk.accounts.dev).
• POSTHOG_API_KEY: API key for server-side analytics tracking.

Tip: You can also set these via the CLI: bunx convex env set AI_GATEWAY_API_KEY your_key_here

3. Clerk Webhook Setup

Clerk webhooks sync user data to Convex. Without this, users will hit an infinite redirect loop on their first sign-in.

Local Development (Tunnel Required)

Since Clerk needs to reach your local server, set up a tunnel:

1. Start a tunnel (choose one):

   # Using ngrok
   ngrok http 3000
   
   # Using cloudflared
   cloudflared tunnel --url http://localhost:3000
   

2. Configure webhook in Clerk Dashboard:

   • Go to Clerk Dashboard → Webhooks → Add Endpoint
   • Set endpoint URL: https://your-tunnel-url.ngrok.io/api/webhooks/clerk
   • Subscribe to events: user.created, user.updated, user.deleted
   • Copy the Signing Secret

3. Add to .env.local:

   CLERK_WEBHOOK_SECRET=whsec_your_signing_secret_here
   CLERK_ISSUER_DOMAIN=your-clerk-frontend-url.clerk.accounts.dev
   

   Important: CLERK_ISSUER_DOMAIN is your Clerk Frontend API URL (found in Clerk Dashboard → API Keys → Frontend API). Format: your-app-name.clerk.accounts.dev for development, clerk.yourdomain.com for production.

Tip: Use a static ngrok URL (free tier includes one) to avoid reconfiguring the webhook each session:

ngrok http 3000 --domain=your-static-subdomain.ngrok-free.app

Set this up once in the Clerk dashboard and you're done.

4. Clerk JWT Template Setup

Required for Convex integration: Configure Clerk to generate JWTs for Convex.

1. Create JWT Template:

   • Go to Clerk Dashboard → JWT templates
   • Click New template → Select Convex
   • Copy the Issuer URL (this matches your CLERK_FRONTEND_API_URL)

2. Verify Claims (pre-configured for Convex):

   • aud: Convex audience (auto-set)
   • name: User's full name from user.full_name
   • Add any additional claims as needed using shortcodes

3. Configure Convex Auth: Your convex/auth.config.ts should reference the environment variable:

   export default {
     providers: [
       {
         domain: process.env.CLERK_ISSUER_DOMAIN,
         applicationID: "convex",
       },
     ],
   };
   

Note: CLERK_ISSUER_DOMAIN should be set without protocol (e.g., your-app-name.clerk.accounts.dev, not https://your-app-name.clerk.accounts.dev).

Reference: For complete Convex + Clerk integration guide, see Clerk Documentation.

Note: The applicationID: 'convex' is correct - it's a constant identifier, not your deployment name.

5. Admin Access Setup (Optional)

To access the admin dashboard (/admin):

1. Set yourself as admin in Convex Dashboard → Data → users table → set isAdmin: true
2. Sync to Clerk:

   bun run scripts/sync-admin-to-clerk.ts
   

3. Sign out and back in to refresh your session token.

Future admin changes via /admin/users auto-sync to Clerk.

6. Seed the Database with Models

Required for new deployments: The models table must be seeded before the app can function.

Run the seed command via Convex dashboard or CLI:

# Via Convex CLI (recommended)
bunx convex run models/seed:seedModels

# Or with clear existing (resets all model data)
bunx convex run models/seed:seedModels '{"clearExisting": true}'

This inserts:

• 40+ AI models (GPT-5, Claude, Gemini, etc.)
• Model profiles for auto-router scoring
• Default auto-router configuration

When to seed:

• Initial deployment (required)
• After bunx convex deploy --reset
• When new models are added to packages/backend/convex/models/seed.ts

7. Run Locally

1. Install dependencies:

   bun install
   

2. Start the Convex backend (in a separate terminal):

   bunx convex dev
   

3. Start the Next.js development server:

   bun dev
   

Open http://localhost:3000 to see the app.