Skip to main content
BenchLM
Data

Kimi Hallucination Benchmark Hard with tools (Internal) (Kimi Hallu Bench Hard (tools))

We show this table for reference; we do not rank on it.

Data verified

The hard split of Moonshot AI's internal factuality benchmark, run with tools and scored as the claim-level hallucination rate.

Benchmark score on Kimi Hallu Bench Hard (tools) — October 29, 2026

We compile the Kimi Hallu Bench Hard (tools) rows from provider self-reports. Kimi K3.1 Pro leads the table at 8.3%. We do not use these results to rank models overall.

1 modelAgenticCurrentDisplay onlyUpdated October 29, 2026

Benchmark score table (1 model)

Benchmark score results for Kimi Hallu Bench Hard (tools)
RankModel / configurationScoreParameters (B)Open / closed
1Kimi K3.1 ProMoonshot AI
8.3%
Not reportedPending

About Kimi Hallu Bench Hard (tools)

Year

2026

Tasks

Hard internal work-task prompts with tools

Format

Claim-level hallucination rate

Difficulty

Factuality stress test

Fact-checking agents judge each factual claim; lower is better. Moonshot describes the set as a deliberately adversarial stress test. BenchLM stores the provider-published internal benchmark value as display-only launch evidence and does not treat it as an independently reproducible score.

Freshness and provenance

Version

Kimi Hallu Bench Hard (tools) 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does Kimi Hallu Bench Hard (tools) measure?

The hard split of Moonshot AI's internal factuality benchmark, run with tools and scored as the claim-level hallucination rate.

Which model scores highest on Kimi Hallu Bench Hard (tools)?

Kimi K3.1 Pro by Moonshot AI currently leads with a score of 8.3% on Kimi Hallu Bench Hard (tools).

How many models are evaluated on Kimi Hallu Bench Hard (tools)?

1 AI models have published results on Kimi Hallu Bench Hard (tools) in the BenchLM catalog.

Last updated: October 29, 2026 · BenchLM version Kimi Hallu Bench Hard (tools) 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 6,100+ subscribers.

One email each week. Unsubscribe anytime.