AI Potluck
Product / UX / Agent tools & protocols

Tavily Search API

Tavily

Search API built for AI agents, returning clean structured results optimized for model consumption rather than human browsing, across search, extract, crawl and map endpoints. It is integrated into LangChain, CrewAI and AutoGen out of the box.

Nebius announced an agreement to acquire Tavily, which is an agreement rather than a closed deal; the adoption score treats it as corroboration and takes its band from the published developer count instead. Verified 2026-08-13 via docs.tavily.com and the Tavily homepage.

Openness

1 high confidence
1.0
license
Proprietary(API-only)
source
closed(no self-hostable core
managed
cloud SaaS with API keys, credits, rate limits

A closed search and scraping API. Only the client SDKs are open; the search service itself is proprietary, with no self-hostable core.

  • https://docs.tavily.com/ recorded 2026-08-13

    Hosted Search, Extract, Crawl, Map and Research endpoints behind an API key. Developer Resources lead with "API Credits Overview - Learn how Tavily API credits work" and "Rate Limits - Understand Tavily's rate limits and policies". The navigation's Open Source group is "popular open source projects that showcase Tavily's use cases", i.e. consumers of the API. No self-hosting or source-available option documented.

Adoption

4 high confidence
4.0

Tavily's own homepage states 2M+ developers and 300M+ monthly requests, billions of pages crawled; it is the default search provider wired into LangChain/CrewAI/LlamaIndex agent stacks. Developer reach solidly in 1-10M band; request volume is far higher. Acquisition by Nebius corroborates enterprise traction. The band is taken on the developer count, which is what puts it at 1M-10M; the request figure would band higher, but requests are not the quantity this scale measures, so it stands as context rather than as the basis.

Capability

4 medium confidence
4.0

A full agent-oriented feature set - coverage, freshness, structured output and published rate limits - purpose-built for agents, and the de facto default in the major agent frameworks. Near-frontier among agent search APIs, but not the single product that defines the category.

  • https://docs.tavily.com/ recorded 2026-08-13

    Search, Extract, Crawl, Map and Research endpoints; a Research API example returning a JSON schema with `citation_format - numbered`; Credits and Rate Limits documented as developer resources.

Verified 2026-08-13