gptme with AI Gateway

gptme is an open-source, terminal-based AI coding agent with shell, file, and browser tools. It supports custom providers through its TOML config, which you can point at AI Gateway for unified model access and spend monitoring.

Run the Vercel CLI setup command for gptme:

npx vercel ai-gateway setup --agent gptme

The command upserts a vercel-ai-gateway entry into the [[providers]] array of ~/.config/gptme/config.toml with api_key_env naming AI_GATEWAY_API_KEY, and points [models].default at the gateway's default model.

If you can't use the Vercel CLI, configure gptme manually:

1. Create an API key

   Go to the AI Gateway section in the Vercel dashboard sidebar and click API keys to create a new API key.

2. Export the key

   Export the key in your shell:

   Terminal

   export AI_GATEWAY_API_KEY="your-ai-gateway-api-key"

   To persist the key, add the same line to your shell's startup file.

3. Add the gateway provider

   Add a [[providers]] entry to ~/.config/gptme/config.toml:

   ~/.config/gptme/config.toml

   [[providers]]
   name = "vercel-ai-gateway"
   base_url = "https://ai-gateway.vercel.sh/coding-agent/v1"
   api_key_env = "AI_GATEWAY_API_KEY"
   default_model = "openai/gpt-6-astra"
    
   [models]
   default = "vercel-ai-gateway/openai/gpt-6-astra"

   api_key_env keeps the key in the shell environment, so it never lands in the config file.

4. Start gptme

   Open a new terminal so the key is loaded, then run:

   Terminal

   gptme

   Run other gateway models by name:

   Terminal

   gptme -m vercel-ai-gateway/anthropic/claude-sonnet-5

View your usage, spend, and request activity in the AI Gateway section in the Vercel dashboard sidebar. See the observability documentation for more details.

• Browse the models catalog for available models
• Configure spend limits on your API key
• Read the gptme documentation for tools and usage