Skip to content
aiqbench

MiniMax · model passport

MiniMax: MiniMax M2

minimax-m2
  • Open weights
  • Apache 2.0 licence
  • 228.7B params
  • 204,800 context
  • Released 2025-10-27(curated)
11/27axes covered
Overall · LMArena
1,343#209 of 396
Best percentile
93.1LiveCodeBench (Artificial Analysis)
Cohort standing
51.6median 53.4

What the data says

Benchmarked on 11 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 93.1 pct — LiveCodeBench (Artificial Analysis) (82.6%)
  • 86.4 pct — IFBench (Artificial Analysis) (72.3%)

Weakest axes

  • 4.7 pct — Factuality Arena (1,360)

Uneven: 88.4 percentile points between its best and worst axis — 93.1 pct on LiveCodeBench (Artificial Analysis) versus 4.7 pct on Factuality Arena.

  • LiveCodeBench (Artificial Analysis)93.1 pct82.6%Verify →
  • IFBench (Artificial Analysis)86.4 pct72.3%Verify →
  • MMLU-Pro (Artificial Analysis)78.7 pct82.0%Verify →
  • LiveBench71.1 pct64.3%Verify →
  • Hard prompts49.7 pct1,367Verify →
  • Instruction following49 pct1,333Verify →
  • Overall47.5 pct1,343Verify →
  • Coding45.5 pct1,382Verify →
  • Terminal-Bench29.5 pct30.0%Verify →
  • WebDev Arena17.3 pct1,297Verify →
  • Factuality Arena4.7 pct1,360Verify →
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

RepositoryMiniMaxAI/MiniMax-M2
Parameters228.7B
ArchitectureMiniMaxM2ForCausalLM
LicenceApache 2.0

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "MiniMaxAI/MiniMax-M2"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Use it hosted

API model IDminimax/minimax-m2
Context length204,800 tokens
Max output131,072 tokens
Input price$0.26 per 1M tokens
Output price$1.02 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningLogprobsMax tokensPresence penaltyReasoningRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop logprobsTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax/minimax-m2","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • minimax-m2 stands at 51.6 (raw mean 52) — measured on 11 of 27 axes.

    standingtemplateVerify →

  • minimax-m2 stands at 51.6 versus a cohort median of 53.4 across 132 models released within 180 days — below the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →