Introducing Parallel Decisions API: the first decision engine with world knowledge built in. Learn more.[Learn more.]
Search Capability Leaderboard

Kimi K3.1 Pro

Moonshot AIClosed

Evaluated

#5 in Search Intelligence / #15 in Search Efficiency

Search Intelligence Score

74.2

+35.5 lift from search

050100
Lift from search
+35.5
38.8 without search
Cost per 1K tasks
$268
#15 in Search Efficiency

Accuracy by benchmark

Accuracy on each eval, with search (solid) and without (light).

  • DSQA91.6 vs 51.3
  • HLE56.0 vs 47.0
  • WISER75.0 vs 18.0

Cost per 1K tasks

How the cost splits between model inference, search calls, and page extraction.

Inference
$252
94%
Search
$5.65
2%
Extract
$10.6
4%

Compared with

Other models from the same provider alongside models that score closest.

Search Intelligence Score averages accuracy on DSQA, HLE, and WISER with equal weight; cost is averaged the same way, with cost shown per 1,000 tasks. Evaluated . Leaderboard last updated . Full methodology.

# Kimi K3.1 Pro / Search Capability Leaderboard

  • Search Intelligence Score: 74.2 (#5 in Search Intelligence)
  • Without search: 38.8 / Lift from search +35.5
  • Cost per 1K tasks: $268 (#15 in Search Efficiency)
  • Evaluated:
  • Lab: Moonshot AI / Weights: Closed / Model id: moonshotai/kimi-k3.1-pro

## Accuracy by benchmark

SuiteWith searchWithout search
DSQA91.651.3
HLE56.047.0
WISER75.018.0

## Compared with

ModelScoreLiftCost per 1K tasks
Kimi K3.1 Pro74.2+35.5$268
Kimi K364.2+33.6$266
Gemini 4 Carbon74.2+33.6$472
Claude Sonnet 5.574.1+40.9$796
Claude Fable 5.173.4+26.5$1,762

Leaderboard last updated . Methodology: /leaderboard#methodology