docketrouter
Models / Moonshot AI

Kimi K2 0711

by Moonshot AI · moonshotai/kimi-k2

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...

tool-usereleased 2025-07-11
Legal score · raw
-
not yet benchmarked
Context
131K
max output 100K
Input
$0.57
per 1M tokens
Output
$2.30
per 1M tokens
Suite cost
-
run the suite to see

Benchmark results

TaskCategoryRawJuicedCorrectLatencyCostRan
Hearsay IdentificationEvidence------
Bluebook Citation FormatResearch & Writing------
Federal Civil ProcedureProcedure------
Limitations ArithmeticProcedure------
Contract Clause ClassificationContracts------
Citation Hallucination ResistanceReliability------

Measured by DocketBuster

These numbers come from DocketBuster's own legal battery, not from DocketRouter's suite. Latest run per metric, with n and a 95% Wilson interval where the source reports one. See docketbuster.com/benchmarks.

MetricValuenIntervalMeasured
Statute pinpoint, exact section (no retrieval)0.3% (1/307)30795% CI 0.1% to 1.8%2026-08-24
Statute pinpoint, exact section (with DocketBuster retrieval)33.2% (102/307)30795% CI 28.2% to 38.7%2026-08-24
Say-nothing rate (declines to bluff when the answer is not in the record)100.0%60095% CI 99.4% to 100.0%2026-08-25
Abstained on statute pinpoint65.1% (200/307)307count, no interval reported2026-08-24

Source files: hard-llm-level-kimi-k26-statute.json, hard-llm-level-kimi-k26-statute_rag.json, hard-llm-level-kimi-k26-saynothing.json. Raw model name in source: moonshotai/kimi-k2.6.