docketrouter
Models / MiniMax

MiniMax M2

by MiniMax · minimax/minimax-m2

MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier intelligence across general reasoning,...

reasoningtool-usereleased 2025-10-23
Legal score · raw
-
not yet benchmarked
Context
205K
max output 131K
Input
$0.26
per 1M tokens
Output
$1.02
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)5.9% (18/307)30795% CI 3.7% to 9.1%2026-08-24
Statute pinpoint, exact section (with DocketBuster retrieval)75.9% (233/307)30795% CI 70.8% to 80.3%2026-08-24
Say-nothing rate (declines to bluff when the answer is not in the record)97.3%60095% CI 95.7% to 98.4%2026-08-25
Abstained on statute pinpoint18.9% (58/307)307count, no interval reported2026-08-24
Coaching quality (GW-14x, 0 to 8)7.28 / 8-rubric mean, no interval reported2026-08-24

Source files: hard-llm-level-minimax-m27-statute.json, hard-llm-level-minimax-m27-statute_rag.json, gw14x-ortier-minimaxm27.json, hard-llm-level-minimax-m27-saynothing.json. Raw model name in source: minimax/minimax-m2.7.