Setting Up Your LLM API Key

SmartDocs works with any LLM provider. Here's how to get started with the most popular ones.

OpenAI (Recommended)

OpenAI is the easiest to set up and works great with gpt-4o-mini (fast and cheap).

1. Get an API Key

Go to platform.openai.com/api-keys
Sign up or log in
Click "Create new secret key"
Give it a name (e.g., "smartdocs")
Copy the key (starts with sk-proj-...)

💰 Cost: gpt-4o-mini is ~$0.15 per 1M input tokens. A typical documentation query costs less than $0.001. Add $5 credit to get started.

2. Configure SmartDocs

SMARTDOCS_AI_PROVIDER=openai
SMARTDOCS_API_KEY=sk-proj-your-key-here
SMARTDOCS_MODEL=gpt-4o-mini

3. Available Models

Model	Best for	Cost
`gpt-4o-mini`	Fast, cheap, good enough	Very low
`gpt-4o`	Best quality, complex questions	Moderate
`gpt-5-nano`	Latest, reasoning-focused	Low

Anthropic (Claude)

Claude is excellent for detailed, nuanced documentation answers.

1. Get an API Key

Go to console.anthropic.com
Sign up or log in
Go to API Keys → Create Key
Copy the key (starts with sk-ant-...)

💰 Cost: Claude Sonnet 4 is ~$3 per 1M input tokens. Similar per-query cost to OpenAI.

2. Configure SmartDocs

SMARTDOCS_AI_PROVIDER=anthropic
SMARTDOCS_API_KEY=sk-ant-your-key-here
SMARTDOCS_MODEL=claude-sonnet-4-20250514

3. Available Models

Model	Best for
`claude-sonnet-4-20250514`	Best balance of speed/quality
`claude-opus-4-20250514`	Maximum quality, complex reasoning

Local with Ollama (Free, Private)

Run models entirely on your machine. No API key needed, no data leaves your computer.

1. Install Ollama

# macOS / Linux
curl -fsSL https://ollama.com/install.sh | sh

# Windows
# Download from https://ollama.com/download

2. Pull a Model

ollama pull llama-3.2-3b
# or for better quality:
ollama pull qwen-2.5-7b

3. Start Ollama

ollama serve
# Runs on http://localhost:11434

4. Configure SmartDocs

SMARTDOCS_AI_PROVIDER=openai-compatible
SMARTDOCS_BASE_URL=http://localhost:11434/v1
SMARTDOCS_MODEL=llama-3.2-3b
# No API key needed for local models

💡 Tip: Local models are free and private but slower and less capable than cloud models. Start with OpenAI or Anthropic, switch to local when you're comfortable.

Other Providers

SmartDocs supports any OpenAI-compatible API. This includes:

Groq — Fast inference, free tier available
Together AI — Many open-source models
Perplexity — Search-augmented models
OpenRouter — Unified API for many providers

Configure them like this:

SMARTDOCS_AI_PROVIDER=openai-compatible
SMARTDOCS_BASE_URL=https://api.groq.com/openai/v1
SMARTDOCS_API_KEY=gsk_your_key_here
SMARTDOCS_MODEL=llama-3.2-3b

Security

Your API key is stored in .env and never sent to the browser
All LLM calls happen server-side in a Next.js API route
Local models (Ollama) never leave your machine
The .env file is gitignored — don't commit your keys

Next: Configuration Guide — full environment variable reference