AI Search

AI Search goes beyond a single search query. It runs multiple searches, analyses the results, and synthesises a comprehensive answer. Ideal for research questions, comparisons, and multi-faceted topics.

How multi-step reasoning works:

1. 1Query analysis — the AI breaks down your question into sub-queries and identifies what information is needed
2. 2Multi-engine search — it runs searches across multiple engines (Bing, Google, DuckDuckGo) to get diverse perspectives
3. 3Content extraction — the top results are scraped and parsed for relevant information
4. 4Reasoning and synthesis — the AI analyses all gathered information, identifies patterns, contradictions, and key facts
5. 5Answer generation — a comprehensive answer is generated with inline source citations

This process is similar to how a human researcher would tackle a complex question — gathering information from multiple sources, cross-referencing facts, and synthesising a well-informed answer.

AI Search costs 8 credits per request. This covers:

• Multiple underlying search calls (2-5 searches across different engines)
• Content extraction from top results (scrape calls)
• AI reasoning and synthesis
• Source citation generation

The cost is fixed at 8 credits regardless of how many sub-queries or searches are performed internally. Credits are deducted upfront — if the AI search fails after deduction, credits are refunded.

Successful AI search response:

AI Search using each SDK:

This endpoint can return the following errors: