MiniMax · mit

MiniMax M2

MiniMax M2 by MiniMax appears in 1 source with Reasoning at 33.6. Best read for Code quality, Long context, Low cost.

CreatorMiniMax

Release date2025-10-27

Knowledge cutoffNot published

Context1M tokens

Input price$0.3/M tokens

Output price$1.2/M tokens

Modalitytext

CountryCN

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank49ranking · source_rank

Reasoning33.6reasoning · index_reasoning

Math25.57math · index_math

Coding23.28coding · index_code

Research4.85research · index_search

Writing19.02writing · index_communication

Vision6.07multimodal · index_vision

Tool calling13.93tool_calling · index_tool_calling

Finance29.56domain · index_finance

Legal29.54domain · index_legal

Healthcare29.28domain · index_healthcare

GPQA78 %reasoning · gpqa_score

AIME 202578 %math · aime_2025_score

SWE-bench Verified69.4 %coding · swe_bench_verified_score

Code Arena966.66 %coding · coding_arena_score

Humanity Last Exam12.5 %reasoning · hle_score

BrowseComp44 %research · browsecomp_score

Terminal Bench46.3 %coding · terminal_bench_score

SciCode36 %coding · scicode_score

Context1,000,000 tokenscontext · context

Speed358.51 c/sperformance · throughput

Latency1,663.1 msperformance · latency

Input price0.3 $/Mpricing · input_price

Output price1.2 $/Mpricing · output_price

Parameters230,000,000,000 paramsmodel · params

Evidence

Citations and source overlap

LLM StatsPublic full LLM leaderboard advanced explorer snapshot, including ranking, benchmark indexes, pricing, speed, context, license, and model metadata.Retrieved 2026-06-01

FAQ

How should I read this profile?

Treat this as a source-backed model dossier, not an EvalKit-run verification. The public values are replicated from linked sources and kept source-scoped.

Is MiniMax M2 verified by EvalKit?

No. EvalKit currently shows 0 verified rows until real run evidence exists.

Why can metrics disagree?

Different sources test different tasks, dates, prompts, and aggregation methods. EvalKit keeps those differences visible instead of merging them into a fake universal score.