Web fetch
openrouter:web_fetch
Lets any model read a web page. When the model needs the contents of a URL, OpenRouter fetches it with the configured engine, extracts the readable content, and returns it to the model.
How it works
For reading a specific page or PDF, like a link in the prompt or documentation to summarize.
Your model requests the URL of a page or PDF it needs to read.
OpenRouter fetches it with the configured engine and extracts the readable text.
Your model uses the content in its answer, and can fetch more URLs.
Providers and pricing
The default engine, auto, uses the model provider’s native fetch when the model supports it and falls back to Exa otherwise.
Exa
$0.001 / page
Extract page content with Exa.
Parallel
$0.001 / page
Extract page content with Parallel.
OpenRouter
Free
Direct page fetch, up to 50 per request.
Firecrawl
Your own account
No OpenRouter charge
Scrape pages using your Firecrawl API key.
Anthropic
NativeBuilt-in web fetch on Claude Opus 4.7 and later and Claude Sonnet 5.
Via Anthropic, Claude Platform on AWS, Amazon Bedrock
Standard model token costs apply to every request.
Using this tool
OpenRouter runs the tool during the request and returns the result to the model, so no client-side tool loop is needed.
Supported APIs: Chat Completions, Responses, and Anthropic Messages.
curl https://openrouter.ai/api/v1/chat/completions \
-H "Authorization: Bearer $OPENROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-5.2",
"tools": [
{
"type": "openrouter:web_fetch"
}
],
"messages": [
{
"role": "user",
"content": "Summarize the page at https://openrouter.ai/docs/guides/features/server-tools"
}
]
}'Setup, parameters, and examples in the docs.
Read the integration guide