Liner tested Tavily's public API while comparing search and deep-research providers in August 2026. Its report includes Tavily basic search in an evaluation of fresh answers and describes testing providers' own public endpoints. This is a published benchmark of Tavily, not a named Tavily feature inside Liner's product.
A July 2026 article also reports Liner's own benchmark runs involving Tavily's advanced answer, Tavily Research and basic search. Liner reports Tavily scores of 51 percent on 2WikiMultiHopQA, 48 percent on MuSiQue and 18 percent on FreshQA. Every system received the same questions. Answers counted as correct when they contained the gold answer. FreshQA was scored by GPT-5.4-mini over 100 questions per system. Liner did not publish the individual provider outputs.
Liner API Pricing: $1 per 1,000 Search Requests, Compared
30 Jul 2026 · build.liner.com · Web page
Liner reports its own benchmark runs involving Tavily's advanced answer, Tavily Research and basic search. Answers counted as correct when they contained the gold answer. GPT-5.4-mini scored FreshQA over 100 questions per system.
Liner API Pricing: $1 per 1,000 Search Requests, Compared30 Jul 2026 · build.liner.com · Web pageLiner reports its own benchmark runs involving Tavily's advanced answer, Tavily Research and basic search. Answers counted as correct when they contained the gold answer. GPT-5.4-mini scored FreshQA over 100 questions per system.
Tavily Alternative Comparison for Deep Research APIs
21 Aug 2026 · Liner · Web page
Liner reports running providers' public endpoints and includes Tavily in its benchmark results.
Tavily Alternative Comparison for Deep Research APIs21 Aug 2026 · Liner · Web pageLiner reports running providers' public endpoints and includes Tavily in its benchmark results.
Liner develops AI search, research and writing tools for students, academics and professionals. Its main search product answers questions with citations and supports follow-up exploration. Liner Scholar helps users find and compare academic papers, select citations and prepare literature reviews. Liner Write combines research, drafting and editing for documents such as emails, newsletters and reports, while Liner Finance focuses on investment research. The company also provides APIs and licenses its search technology for integration into enterprise products and internal research workflows.
Liner launched Actions MCP, which lets AI agents connect to about 1,100 external apps through one integration. It also introduced Actions API for developers building that access into their own agents.
Liner signed an MOU with KAIST’s di-Lab to develop AI tools for policy research. The collaboration covers research workflows, knowledge graphs, verification agents and policy simulations.
Liner launches Model API for automatic model routing
Liner launched an API that selects models according to each request’s requirements. The company says it can reduce token costs by more than 50% while preserving response quality.
Liner develops AI search and research tools. · Consumer
How they use Tavily
Liner tested Tavily's public API while comparing search and deep-research providers in August 2026. Its report includes Tavily basic search in an evaluation of fresh answers and describes testing providers' own public endpoints. This is a published benchmark of Tavily, not a named Tavily feature inside Liner's product.
A July 2026 article also reports Liner's own benchmark runs involving Tavily's advanced answer, Tavily Research and basic search. Liner reports Tavily scores of 51 percent on 2WikiMultiHopQA, 48 percent on MuSiQue and 18 percent on FreshQA. Every system received the same questions. Answers counted as correct when they contained the gold answer. FreshQA was scored by GPT-5.4-mini over 100 questions per system. Liner did not publish the individual provider outputs.
Liner API Pricing: $1 per 1,000 Search Requests, Compared
30 Jul 2026 · build.liner.com · Web page
Liner reports its own benchmark runs involving Tavily's advanced answer, Tavily Research and basic search. Answers counted as correct when they contained the gold answer. GPT-5.4-mini scored FreshQA over 100 questions per system.
Liner API Pricing: $1 per 1,000 Search Requests, Compared30 Jul 2026 · build.liner.com · Web pageLiner reports its own benchmark runs involving Tavily's advanced answer, Tavily Research and basic search. Answers counted as correct when they contained the gold answer. GPT-5.4-mini scored FreshQA over 100 questions per system.
Tavily Alternative Comparison for Deep Research APIs
21 Aug 2026 · Liner · Web page
Liner reports running providers' public endpoints and includes Tavily in its benchmark results.
Tavily Alternative Comparison for Deep Research APIs21 Aug 2026 · Liner · Web pageLiner reports running providers' public endpoints and includes Tavily in its benchmark results.
Liner develops AI search, research and writing tools for students, academics and professionals. Its main search product answers questions with citations and supports follow-up exploration. Liner Scholar helps users find and compare academic papers, select citations and prepare literature reviews. Liner Write combines research, drafting and editing for documents such as emails, newsletters and reports, while Liner Finance focuses on investment research. The company also provides APIs and licenses its search technology for integration into enterprise products and internal research workflows.
Liner launched Actions MCP, which lets AI agents connect to about 1,100 external apps through one integration. It also introduced Actions API for developers building that access into their own agents.
Liner signed an MOU with KAIST’s di-Lab to develop AI tools for policy research. The collaboration covers research workflows, knowledge graphs, verification agents and policy simulations.
Liner launches Model API for automatic model routing
Liner launched an API that selects models according to each request’s requirements. The company says it can reduce token costs by more than 50% while preserving response quality.
