MiniMax

MiniMax-M2.1

DeprecatedOpen weights205K contextReasoning

Summary

MiniMax-M2.1 is an open-weight language model from MiniMax, released on 23 Dec 2025. It accepts text and generates text, with a 205K-token context window and up to 131K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.30 per million input tokens and $1.20 per million output tokens (Hugging Face Inference Providers), across 2 providers we track. It is scheduled for retirement on 5 Nov 2026.

Earlier MiniMax agent model for practical coding and productivity tasks

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperMiniMax
Release date23 Dec 2025 [models.dev]
StatusDeprecated [models.dev]
WeightsOpen weights [Hugging Face]
Licensemodified-mit [Hugging Face]
Parameters229B [Hugging Face]
Context window205K (204,800 tokens) [models.dev]
Max output131K tokens [models.dev]
Knowledge cutoffJan 2025 [models.dev]
InputText [models.dev]
OutputText [models.dev]
Reasoning modeYes [models.dev]
Tool callingYes [models.dev]
Retirement date5 Nov 2026 [OpenRouter]
Hugging FaceMiniMaxAI/MiniMax-M2.1 [Hugging Face]

MiniMax-M2.1 API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
Hugging Face Inference Providers
MiniMaxAI/MiniMax-M2.1
$0.30$1.20–205Kmodels.dev, 18 Nov 2026
Vercel AI Gateway
minimax/minimax-m2.1
$0.30$1.20$0.03205Kmodels.dev, 18 Nov 2026

Running MiniMax-M2.1 locally

Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.

QuantizationWeightsTotal, 8K contextTotal, 32K context
Q4_K_M (4-bit)139 GB142 GB148 GB
Q8_0 (8-bit)243 GB246 GB252 GB
FP16 / BF16 (16-bit)457 GB460 GB467 GB

At 4-bit with an 8K conversation, MiniMax-M2.1 needs about 142 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.

Frequently asked questions

What is the context window of MiniMax-M2.1?

MiniMax-M2.1 has a context window of 204,800 tokens (205K) and can generate up to 131,072 tokens in a single response, according to models.dev.

How much does the MiniMax-M2.1 API cost?

MiniMax-M2.1 is listed from $0.30 per million input tokens and $1.20 per million output tokens (Hugging Face Inference Providers). Prices last verified 18 Nov 2026.

Is MiniMax-M2.1 open source?

MiniMax-M2.1 is an open-weight model: its weights are published on Hugging Face as MiniMaxAI/MiniMax-M2.1 under the modified-mit license. Check the license terms before commercial use.

When was MiniMax-M2.1 released?

MiniMax released MiniMax-M2.1 on 23 Dec 2025, according to models.dev.

Which providers offer MiniMax-M2.1?

We track MiniMax-M2.1 on 2 providers: Hugging Face Inference Providers, Vercel AI Gateway.

How much VRAM does MiniMax-M2.1 need?

About 142 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 246 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
GLM-4.7Zhipu AI (Z.ai)22 Dec 2025205K$0.40$1.75Open
DeepSeek ChatDeepSeek1 Dec 2025164K$0.26$1.03Open
GLM-4.6VZhipu AI (Z.ai)8 Dec 2025128K$0.30$0.90Open
Qwen3.5 35B-A3BAlibaba (Qwen)23 Feb 2026262K$0.14$1Open
DeepSeek V3.1 TerminusDeepSeek22 Sept 2025164K$0.27$1Open
Qwen3 Coder NextAlibaba (Qwen)3 Feb 2026262K$0.12$0.80Open

Change history

  • Knowledge cutoff changed from 1 Jan 2025 to Jan 2025
  • Knowledge cutoff changed from Jan 2025 to 1 Jan 2025
  • Knowledge cutoff changed from Oct 2025 to Jan 2025
  • Knowledge cutoff changed from 1 Jan 2025 to Oct 2025
  • Knowledge cutoff changed from Oct 2025 to 1 Jan 2025
  • Knowledge cutoff changed from Jan 2025 to Oct 2025
  • Knowledge cutoff changed from 1 Jan 2025 to Jan 2025
  • Knowledge cutoff changed from Jan 2025 to 1 Jan 2025
  • Knowledge cutoff changed from Oct 2025 to Jan 2025
  • Knowledge cutoff changed from Jan 2025 to Oct 2025
  • Knowledge cutoff changed from Oct 2025 to Jan 2025
  • No longer listed on openrouter
  • Retirement date changed from 8 Oct 2026 to 5 Nov 2026
  • Knowledge cutoff changed from 1 Jan 2025 to Oct 2025
  • Knowledge cutoff changed from Jan 2025 to 1 Jan 2025
  • Knowledge cutoff changed from Oct 2025 to Jan 2025
  • Knowledge cutoff changed from 1 Jan 2025 to Oct 2025
  • Knowledge cutoff changed from Oct 2025 to 1 Jan 2025
  • Knowledge cutoff changed from 1 Jan 2025 to Oct 2025
  • Knowledge cutoff changed from Oct 2025 to 1 Jan 2025
  • Status changed from active to deprecated
  • No longer listed on amazon-bedrock
  • No longer listed on minimax
  • Knowledge cutoff changed from 1 Jan 2025 to Oct 2025
  • No longer listed on novita-ai

Sources