MiniMax-M2
Summary
MiniMax-M2 is an open-weight language model from MiniMax, released on 27 Oct 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 1 provider we track. It is scheduled for retirement on 5 Nov 2026.
Efficient open MiniMax model built for coding agents and tool-heavy workflows
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | MiniMax |
|---|---|
| Release date | 27 Oct 2025 [models.dev] |
| Status | Deprecated [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | modified-mit [Hugging Face] |
| Parameters | 229B [Hugging Face] |
| Context window | 205K (204,800 tokens) [models.dev] |
| Max output | 131K tokens [models.dev] |
| Knowledge cutoff | 1 Jan 2025 [models.dev] |
| Input | Text [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Retirement date | 5 Nov 2026 [OpenRouter] |
| Hugging Face | MiniMaxAI/MiniMax-M2 [Hugging Face] |
MiniMax-M2 API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| Hugging Face Inference Providers MiniMaxAI/MiniMax-M2 | $0.30 | $1.20 | – | 205K | models.dev, 18 Nov 2026 |
Running MiniMax-M2 locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 139 GB | 142 GB | 148 GB |
| Q8_0 (8-bit) | 243 GB | 246 GB | 252 GB |
| FP16 / BF16 (16-bit) | 457 GB | 460 GB | 467 GB |
At 4-bit with an 8K conversation, MiniMax-M2 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?
MiniMax-M2 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 API cost?
MiniMax-M2 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 open source?
MiniMax-M2 is an open-weight model: its weights are published on Hugging Face as MiniMaxAI/MiniMax-M2 under the modified-mit license. Check the license terms before commercial use.
When was MiniMax-M2 released?
MiniMax released MiniMax-M2 on 27 Oct 2025, according to models.dev.
How much VRAM does MiniMax-M2 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.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| DeepSeek Chat | DeepSeek | 1 Dec 2025 | 164K | $0.26 | $1.03 | Open |
| DeepSeek V3.1 Terminus | DeepSeek | 22 Sept 2025 | 164K | $0.27 | $1 | Open |
| GLM-4.6 | Zhipu AI (Z.ai) | 30 Sept 2025 | 205K | $0.50 | $2 | Open |
| GLM-4.7 | Zhipu AI (Z.ai) | 22 Dec 2025 | 205K | $0.40 | $1.75 | Open |
| Qwen3 VL 8B Instruct | Alibaba (Qwen) | 14 Oct 2025 | 262K | $0.25 | $0.75 | Open |
| Qwen3 VL 235B A22B Instruct | Alibaba (Qwen) | 23 Sept 2025 | 262K | $0.20 | $0.88 | Open |
Change history
- No longer listed on vercel
- No longer listed on openrouter
- Retirement date changed from none to 5 Nov 2026
- No longer listed on minimax
- No longer listed on amazon-bedrock
- Status changed from active to deprecated
- openrouter price changed: input $0.30 → $0.26, output $1.20 → $1.02 per 1M tokens
- No longer listed on novita-ai