Live web answers with citations

Native Web Search

Give models real-time web access with a single web_search tool and get citations back with the answer.

Get startedRead the docs
tools: [{ "type": "web_search" }]
"content": "It is 18°C and sunny in San Francisco.",
"annotations": [
  {
    "type": "url_citation",
    "url_citation": {
      "url": "https://weather.example.com/sf",
      "title": "San Francisco forecast"
    }
  }
]
Answer plus sources
01

One tool

Add web_search, done.

02

Citations

Sources returned with the answer.

03

Streaming

Works with streamed responses.

04

SDK friendly

Works with the OpenAI SDK.

Built for

  • Current events

    Answers past the training cutoff

  • Fact checking

    Show where facts come from

  • Live data

    Prices and weather in chat

Turn on search
terminal
curl https://api.llmgateway.io/v1/chat/completions \  -H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \  -H "Content-Type: application/json" \  -d '{    "model": "gpt-5.2",    "messages": [      {        "role": "user",        "content": "Weather in San Francisco?"      }    ],    "tools": [      {        "type": "web_search"      }    ]  }'
Read docs

More in models & modalities

All features

Video Generation

Async video jobs with signed callbacks

Speech Generation

Text to natural speech

Transcription

Speech to text with word timestamps

Start using Native Web Search today

One API key. Every model. Live in minutes.

Start buildingBrowse models