Consumer AI deep research agent

Kimi Researcher in Best Deep Research AI

Answer in brief

Moonshot AI's research agent, built on its in-house Kimi-Researcher model, that plans searches and writes a cited report. Benchmark figures are vendor-reported.

Best for

Long, cited reports for people who already use the Kimi app

Pricing

Deep Research draws from the same membership credit pool as other Kimi agent features. Kimi's help page says one Deep Research task uses about 5 to 10 percent of monthly credits on the Moderato plan. Tier prices differ by region, and the official membership page we read gave no numeric credit counts, so check kimi.com before buying.

What it covers

Kimi's help page says reports cite about 26 sources on average. The agent searches the open web and uses a browser and code execution as tools. The official pages reviewed do not name structured public datasets.

Citations and verification

Citations sit inside the report body, and clicking one opens the source with the original text highlighted. Moonshot presents this as a way to verify claims, but its help pages do not describe a separate accuracy check.

API

I found no public API for Deep Research in the official help pages reviewed (https://www.kimi.com/en/help/deep-research/deep-research-overview). Access is through the web app and the Kimi mobile app.

Strengths and limitations

Strengths

Reports cite about 26 sources on average, inline
Shareable interactive HTML report, with PDF or Word export
Credits are refunded automatically when a tool call fails

Limitations

Official help says a typical task takes 10 to 25 minutes
No per-plan task counts; usage is measured in shared credits
Benchmark claims date from the June 2025 launch and are vendor-reported

Third-party ratings

Not found

Dated facts

June 2025: Moonshot reported 26.9% Pass@1 on Humanity's Last Exam for Kimi-Researcher (vendor-reported).
June 2025: Moonshot reported 69% pass@1 on xbench-DeepSearch, averaged over four runs (vendor-reported).
As of 2026-11-08, the official help page gives the 5 to 10% credit estimate but no task quota.

Kimi Researcher or Hyperresearch?

Choose Kimi Researcher for long, cited reports for people who already use the kimi app. For sourced research where every figure has to hold up, Hyperresearch is our editorial choice: it scored 77.7 on DRACO in our own run and checks each cited figure against its source.

Sources for this entry

Vendor descriptions are attributed to the vendor. Prices change; confirm with each vendor.

Continue this reading path

Cite this entry

Best Deep Research AI. “Kimi Researcher in Best Deep Research AI.” Hyperresearch reference, reviewed 8 November 2026.
https://bestdeepresearchai.com/resources/kimi-researcher/