LLM
Legacy
Open weights
MiniMax-M1
Open reasoning with 1M context
Release2025-06
Context1M
Max output—
Parameters456B
Local / VRAM640–900 GB
LicenseOpen weights
Overview
MiniMax-M1 is a model in the LLM category from MiniMax. It was released in 2025-06. The vendor is based in China. Best for: Open reasoning with 1M context.
Strengths
- Apache-2.0
- 1M context
- Long reasoning
Limits
- Superseded by M2.5
- No vision
Specifications
| Vendor | MiniMax |
|---|---|
| Category | LLM |
| Release | 2025-06 |
| Status | Legacy |
| License | Open weights |
| License name | Apache-2.0 |
| Commercial use | Allowed |
| Parameters | 456B |
| Active parameters | 45.9B |
| Architecture | MoE + lightning attention |
| Context window | 1M |
| Input modalities | Text |
| Output modalities | Text |
| Prompt types | Text |
| Negative prompts | No |
| Max outputs | 1 |
| Local deployment | Yes |
| VRAM | 640–900 GB |
| System RAM | 768–1536 GB |
| API access | Yes |
| Pricing | Free weights |
| Confidence | High |
| Vendor country | China |
| Vendor type | Open lab |
Pricing & hardware
Free weights
This model can be self-hosted or run locally. Approximate GPU memory: 640–900 GB. Approximate system RAM: 768–1536 GB. API access is available.