Scan to join WeChat groupMiniMax M3
MiniMax M3 for coding, agent workflows and complex production tasks
MiniMax M3 is a MiniMax model for coding, agent workflows and complex production tasks, often evaluated for Chinese assistants, generation and multimedia workflows.
starsCapabilities
visibilityVision understandingcodeFunction callingstreamStreaming output
paymentsContext and pricing
Context limit1,000,000
Max output128,000
Input price$0.3/ 1M tokens
Output price$1.2/ 1M tokens
Cached input price$0.06/ 1M tokens
descriptionOverview
Overview
MiniMax M3 is listed in MiniMax's official platform documentation, with model ID MiniMax-M3. MiniMax models cover conversation, text generation, agents, reasoning and speech, making them useful for Chinese products and multimedia applications.
Best for
Use MiniMax M3 for Chinese customer support, writing assistants, long-text understanding, agent tasks or multimedia interaction. Test output style, context handling, cost, latency and endpoint capabilities before production.
lightbulbUse cases
- Chinese support and business assistants
- Content generation and writing
- Long-text understanding
- Agents and complex task planning
thumb_upStrengths
- Good fit for Chinese product scenarios
- Covers text, reasoning and speech directions
- Supports tiered choices inside MiniMax
- Useful for assistants and content applications
infoLimitations
- Capabilities vary significantly by model endpoint
- Long-text and reasoning tasks need testing
- Speech capabilities should be evaluated separately
- Limits depend on MiniMax documentation