admin-valentin/open-notebook
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
open-notebook/docs/4-AI-PROVIDERS/index.md
LUIS NOVO e13e4a2d8b docs: restructure documentation with new organized layout 
- Replace old docs structure with new comprehensive documentation
- Organize into 8 major sections (0-START-HERE through 7-DEVELOPMENT)
- Convert CONFIGURATION.md, CONTRIBUTING.md, MAINTAINER_GUIDE.md to redirects
- Remove outdated MIGRATION.md and DESIGN_PRINCIPLES.md
- Fix all internal documentation links and cross-references
- Add progressive disclosure paths for different user types
- Include 44 focused guides covering all features
- Update README.md to remove v1.0 breaking changes notice
2026-01-03 20:10:24 -03:00

8.3 KiB
Raw Blame History

AI Providers - Setup & Configuration

Open Notebook supports 15+ AI providers. This section helps you choose and configure yours.

Quick Decision: Which Provider?

Cloud Providers (Easiest)

OpenAI (Recommended)

  • Models: GPT-4o, GPT-4o-mini
  • Cost: ~$0.03-0.15 per 1K tokens
  • Speed: Very fast
  • Setup: 5 minutes
  • Best for: Most users (best quality/price balance)

→ Setup Guide

Anthropic (Claude)

  • Models: Claude 3.5 Sonnet, Haiku, Opus
  • Cost: ~$0.80-3.00 per 1M tokens
  • Speed: Fast
  • Setup: 5 minutes
  • Best for: Long context, reasoning
  • Advantage: 200K token context

→ Setup Guide

Google Gemini

  • Models: Gemini 2.0 Flash, 1.5 Pro
  • Cost: ~$0.075-0.30 per 1K tokens
  • Speed: Very fast
  • Setup: 5 minutes
  • Best for: Multimodal (images, audio, video)
  • Advantage: 1M token context

→ Setup Guide

Groq (Ultra-Fast)

  • Models: Mixtral, Llama 3.3
  • Cost: ~$0.05 per 1M tokens (cheapest)
  • Speed: Ultra-fast (fastest available)
  • Setup: 5 minutes
  • Best for: Budget-conscious, transformations
  • Disadvantage: Limited model selection

→ Setup Guide

OpenRouter (100+ Models)

  • Models: Access to OpenAI, Anthropic, Google, Llama, Mistral, and 100+ more
  • Cost: Pay-per-model (varies)
  • Speed: Varies by model
  • Setup: 5 minutes
  • Best for: Model comparison, testing many models, unified billing
  • Advantage: One API key for all models

→ Setup Guide

Local / Self-Hosted (Free)

Ollama (Recommended for Local)

  • Models: Mistral, Llama 2, Phi, Neural Chat
  • Cost: Free (electricity only)
  • Speed: Depends on hardware (slow on CPU, fast on GPU)
  • Setup: 10 minutes
  • Best for: Privacy-first, offline use
  • Privacy: 100% local, nothing leaves your machine

→ Setup Guide

LM Studio (Alternative)

  • GUI-based local LLM runner
  • Cost: Free
  • Speed: Depends on hardware
  • Setup: 15 minutes (GUI easier than Ollama CLI)
  • Best for: Non-technical users
  • Privacy: 100% local

→ Setup Guide

Enterprise

Azure OpenAI

  • Same as OpenAI but on Azure
  • Cost: Same as OpenAI
  • Setup: 10 minutes (more complex)
  • Best for: Enterprise, compliance (HIPAA, SOC2)

→ Setup Guide

Comparison Table

Provider Speed Cost Quality Privacy Setup Models
OpenAI Very Fast Excellent Low 5 min Many
Anthropic Fast Excellent Low 5 min Few
Google Very Fast Good Low 5 min Few
Groq Ultra Fast $ Good Low 5 min Few
OpenRouter Varies Varies Varies Low 5 min 100+
Ollama Slow-Medium Free Good Max 10 min Many
LM Studio Slow-Medium Free Good Max 15 min Many
Azure Very Fast Excellent High 10 min Many

Choosing Your Provider

I want the easiest setup

→ OpenAI — Most popular, best community support

I have unlimited budget

→ OpenAI — Best quality

I want to save money

→ Groq — Cheapest cloud ($0.05 per 1M tokens)

I want privacy/offline

→ Ollama — Free, local, private

I want a GUI (not CLI)

→ LM Studio — Desktop app

I'm in an enterprise

→ Azure OpenAI — Compliance, support

I need long context (200K+ tokens)

→ Anthropic — Best long-context model

I need multimodal (images, audio, video)

→ Google Gemini — Best multimodal support

I want access to many models with one API key

→ OpenRouter — 100+ models, unified billing

Setup Paths

Path 1: OpenAI (Most Common)

1. Go to https://platform.openai.com/api-keys
2. Create account, add $5+ credit
3. Create API key
4. Add to .env: OPENAI_API_KEY=sk-...
5. Restart services
6. Done!

Time: 5 minutes Cost: Pay as you go (~$1-5/month light use)

Path 2: Local Ollama (Privacy)

1. Download Ollama: https://ollama.ai
2. Run: ollama serve
3. Download model: ollama pull mistral
4. Add to .env: OLLAMA_API_BASE=http://localhost:11434
5. Restart services
6. Done!

Time: 10 minutes Cost: Free (electricity only) Requirement: 8GB RAM minimum (16GB+ recommended)

Path 3: Anthropic (Better Reasoning)

1. Go to https://console.anthropic.com/
2. Create account, add payment method
3. Create API key
4. Add to .env: ANTHROPIC_API_KEY=sk-ant-...
5. Restart services
6. Done!

Time: 5 minutes Cost: Pay as you go (~$2-20/month typical use)

Path 4: OpenRouter (100+ Models)

1. Go to https://openrouter.ai/keys
2. Create account, add credit
3. Create API key
4. Add to .env: OPENROUTER_API_KEY=sk-or-...
5. Restart services
6. Done!

Time: 5 minutes Cost: Varies by model ($0.05-15 per 1M tokens)

Model Recommendations by Task

General Chat

  • OpenAI: gpt-4o (best) or gpt-4o-mini (cheap)
  • Anthropic: claude-3-5-sonnet (best)
  • Google: gemini-2.0-flash (balanced)
  • Groq: mixtral-8x7b-32768 (fast)
  • Ollama: mistral (balanced)

Long Documents (200K+ tokens)

  • Anthropic: claude-3-5-sonnet (200K context)
  • Google: gemini-1.5-pro (1M context)
  • OpenAI: gpt-4-turbo (128K context)

Fast/Cheap Operations

  • Groq: Ultra-fast, very cheap
  • OpenAI: gpt-4o-mini (fast, cheap)
  • Anthropic: claude-3-5-haiku (cheap)

Multimodal (Images/Audio/Video)

  • Google: gemini-2.0-flash (best multimodal)
  • OpenAI: gpt-4o (good multimodal)

Local/Offline

  • Ollama: mistral (best local balance)
  • LM Studio: mistral or llama2

Complete Configuration Reference

For detailed environment variable setup, see: → Complete Configuration Reference

For detailed AI provider configuration, see: → AI Providers Configuration

Testing Your Setup

Once configured:

1. Start Open Notebook
2. Go to Settings → Models
3. Select your configured provider
4. Try a Chat question
5. If it responds, you're good!

Cost Estimator

OpenAI

Light use (10 chats/day): $1-5/month
Medium use (50 chats/day): $10-30/month
Heavy use (all-day use): $50-100+/month

Anthropic

Light use: $1-3/month
Medium use: $5-20/month
Heavy use: $20-50+/month

Groq

Light use: $0-1/month
Medium use: $2-5/month
Heavy use: $5-20/month

Ollama

Any use: Free (electricity only)
8GB GPU running 24/7: ~$10/month electricity

Switching Providers

You can switch providers anytime:

1. Edit .env
2. Change/add API key
3. Restart services
4. Go to Settings → Models
5. Select new provider
6. Done!

Your existing notebooks and data stay the same.

Provider-Specific Guides

For detailed setup instructions per provider, see: → 5-CONFIGURATION/ai-providers.md

Includes:

  • OpenAI
  • Anthropic
  • Google Gemini
  • Groq
  • OpenRouter
  • Mistral
  • DeepSeek
  • xAI
  • Ollama
  • LM Studio
  • OpenAI-Compatible
  • Azure OpenAI

Troubleshooting

"Models not showing in Settings"

  • API key missing or wrong
  • Restart services after changing .env
  • Check the key is set correctly: echo $OPENAI_API_KEY

"API key invalid"

  • Copy fresh key from provider's dashboard
  • Check no extra spaces: OPENAI_API_KEY="sk-..."
  • Verify key format matches provider

"Rate limit exceeded"

  • You're hitting provider's rate limits
  • Wait a bit, retry
  • For cloud: upgrade account or reduce request rate

"Connection refused" (Ollama)

  • Ollama not running: ollama serve
  • Wrong port: Check OLLAMA_API_BASE
  • Not on localhost: Use IP instead

Next Steps

  1. Pick a provider from above
  2. Follow setup guide in 5-CONFIGURATION/ai-providers.md
  3. Add API key to .env
  4. Restart services
  5. Test in Settings → Models
  6. Start using!

Support