Inference
EN
MiniMax

MiniMax M2.1

MiniMax M2.1 is a model from MiniMax. 9 benchmarks count toward its index, in 4 categories.

  • Open weights
  • Input: text
  • Released 2025-12
  • Retires 5 Nov 2026
RankedEnough results for an index.
IndexAll categories, out of 100.49.1±8.9Median 51.0
CoverageIndex weight that has results.65%
SpeedOutput tokens per second.34/sMedian 67/s
First tokenSeconds to the first output token.1.7 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.30Median $0.60
Output / 1MUS dollars per 1M output tokens.$1.20Median $2.55
ContextMaximum tokens in one request.205KMedian 262K
EloLMArena rating and rank.—Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
minimax/minimax-m2.1
Max outputMaximum output tokens per request.
131K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
No
Cache read / 1MUS dollars per 1M cached tokens.
$0.03
WeightsLink to the published weights.
Hugging Face (opens in a new tab) ↗
RetiresDate OpenRouter removes the model.
5 Nov 2026

Run locally

228.69B parameters
16-bit
481.2 GB
8-bit
256.1 GB
4-bit
146.6 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy needs more than 128 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.2 counted48.5141 / 218
CodingCode writing and repair.3 counted51.6127 / 209
ReasoningLogic problems and puzzles.None counted——
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.2 counted51.7123 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.2 counted43.9110 / 168

Results

9 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MMLU ProKnowledge87.0% ±0.335 / 135
Terminal-Bench 1.0Agentic41.3% ±5.518 / 47
LiveCodeBenchCoding81.8% ±1.159 / 142
MATH 500Math89.0% ±1.428 / 59
AIMEMath77.9% ±1.349 / 95
GPQA Diamond · Vals AIKnowledge78.5% ±2.172 / 135
SWE-benchCoding74.8% ±1.947 / 87
Terminal-Bench 2.0 · Vals AIAgentic37.1% ±5.138 / 66
IOI v1Coding2.3% ±1.356 / 61
MedQAHealthcare91.2% ±0.340 / 95
MedScribeHealthcare80.8% ±1.855 / 109
CaseLaw v2Legal55.8% ±0.532 / 53
CorpFin v2Finance58.9% ±1.081 / 131
MGSMMultilingual87.9% ±1.856 / 74
MedCodeHealthcare34.1% ±1.981 / 106
LegalBenchLegal75.4% ±0.5115 / 149
TaxEval v2Finance66.4% ±0.9112 / 143
Finance Agent v1.1Finance33.4% ±2.644 / 51

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseVals AINo licence stated

History

5 releases
Index50.2 → 49.1
23 Oct 202612 Nov 2026
Output / 1M$1.20 → $1.20
23 Oct 202612 Nov 2026

Questions

How much does MiniMax M2.1 cost?
$0.30 per 1M input tokens, $1.20 per 1M output tokens.
How good is MiniMax M2.1?
Its index is 49.1 out of 100. That is place 155 of 283 ranked models.
What is MiniMax M2.1 best at?
Knowledge: 51.7 out of 100, place 123 of 268.
How fast is MiniMax M2.1?
34 output tokens per second, 1.7 s to the first token.
What is the context window of MiniMax M2.1?
205K tokens.
Is MiniMax M2.1 open source?
Yes. Its weights are published.
When was MiniMax M2.1 released?
December 2025.

Badge

MiniMax M2.1 on Inference360: place 155 of 283, index 49.1
HTML
<a href="https://inference360.ai/models/minimax-minimax-m2-1"><img src="https://inference360.ai/badges/minimax-minimax-m2-1.svg" alt="MiniMax M2.1 on Inference360: place 155 of 283, index 49.1" height="20"></a>
Markdown
[![MiniMax M2.1 on Inference360: place 155 of 283, index 49.1](https://inference360.ai/badges/minimax-minimax-m2-1.svg)](https://inference360.ai/models/minimax-minimax-m2-1)

Compare with

DeepSeek V3.2 (Thinking)49.4MiniMax M2.550.2

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from MiniMax

All models
MiniMax M2-herUnrankedMiniMax M2.550.2MiniMax M2UnrankedMiniMax M2.751.2MiniMax M355.7