MiniMax M1 is an open-weight reasoning model with a 1M-token context window and hybrid mixture-of-experts architecture designed for efficient inference at scale.
- Strengths
- Handles very long documents and extended context windows efficiently through its mixture-of-experts design and optimized attention mechanism.
- Best for
- Applications requiring processing of large documents, long-form reasoning, and inference efficiency where extended context is needed.
- Limitations
- As an open-weight model, it may require self-hosting and tuning; performance on specialized reasoning tasks may lag behind larger proprietary models.
Input / 1M
$0.55
Output / 1M
$2.20
Cached input / 1M
—
Context window
1M
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in increased 37.5% out decreased 0.0%
- 90d
- in increased 37.5% out decreased 0.0%
- 1y
- in increased 37.5% out decreased 0.0%
- Since launch
- in increased 37.5% out decreased 0.0%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 15 Jul 2026 | $0.55 | $2.20 | — | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.4 | $2.20 | — | Imported from OpenRouter | openrouter.ai |