← All models
MiniMax · mit
MiniMax M2.1
MiniMax M2.1 by MiniMax appears in 1 source with Reasoning at 41.25. Best read for Code quality, Long context, Low cost.
CreatorMiniMax
Release date2025-12-23
Knowledge cutoffNot published
Context1M tokens
Input price$0.3/M tokens
Output price$1.2/M tokens
Modalitytext
CountryCN
Metrics
All source-backed metrics
LLM Stats Rank59ranking · source_rank
Reasoning41.25reasoning · index_reasoning
Math36.17math · index_math
Coding26.94coding · index_code
Research19.1research · index_search
Writing18.17writing · index_communication
Vision17.86multimodal · index_vision
Tool calling20.34tool_calling · index_tool_calling
Long context19.95long_context · index_long_context
Finance49.93domain · index_finance
Legal49.92domain · index_legal
Healthcare49.7domain · index_healthcare
GPQA81 %reasoning · gpqa_score
AIME 202581 %math · aime_2025_score
SWE-bench Verified67 %coding · swe_bench_verified_score
Code Arena861.17 %coding · coding_arena_score
Humanity Last Exam22 %reasoning · hle_score
BrowseComp62 %research · browsecomp_score
Toolathlon43.5 %tool_calling · toolathlon_score
Terminal Bench47.9 %coding · terminal_bench_score
SciCode39 %coding · scicode_score
Context1,000,000 tokenscontext · context
Speed371.86 c/sperformance · throughput
Latency2,193.4 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
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 M2.1 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.