minimax/minimax-m3

minimax/minimax-m3 is a efficient-tier language model from RouterPlus. minimax/minimax-m3 is a proprietary model available exclusively through RouterPlus's API. The MiniMax M3 family is RouterPlus's model series for this capability tier.

minimax/minimax-m3 is priced at $0.28/1M input tokens and $1.1/1M output tokens. Output tokens are 3.9× more expensive than input tokens — this ratio reflects the sequential nature of autoregressive generation, which requires a separate forward pass per output token versus parallel processing of all input tokens simultaneously. At a typical usage pattern of 1,000 input tokens and 500 output tokens per call, minimax/minimax-m3 costs approximately $0.83 per 1,000 API calls. Prompt caching reduces repeated input tokens to $0.056/1M — a 80% discount that compounds significantly for applications with long shared system prompts, multi-turn conversations, or document Q&A where the same context is reused across many queries.

Its 524K-token context window is in the top tier for production models, enabling full-document analysis, long multi-turn conversations, and large codebase understanding without chunking. Maximum output length is 131K tokens — sufficient for long-form content generation, detailed code files, and extended reports.

minimax/minimax-m3 is best suited to general language tasks. Supported capabilities include streaming token output for real-time UIs. When evaluating this model for a specific workload, the most important factors are: context window (does your input fit?), output price (how long are your generations?), and latency (is this interactive or batch?). The pricing table above shows how minimax/minimax-m3 compares to similar models across these dimensions.

minimax/minimax-m3 sits in the efficient tier — optimized for speed and cost while maintaining solid quality, ideal for high-volume or latency-sensitive workloads. Unlike open-weights alternatives, minimax/minimax-m3 is only available via RouterPlus's API — there is no self-hosted option, which simplifies deployment but creates vendor dependency. When comparing minimax/minimax-m3 to alternatives, use the similar models section on this page to find models at comparable price points with different capability tradeoffs. The intelligence score (N/A/100) reflects aggregate benchmark performance — a useful proxy for general capability, though real-world performance varies significantly by task type. Always test with your specific prompts and use case before committing to a model for production.