messages
Required
Type: JSON array
The conversation sent to the chat model in chronological order. Each JSON item contains a role and its text content.
China64K contextRole-playingText
Explore MiniMax M2-her
Create character-rich role-playing and long-running conversations with explicit model roles, user personas, example dialogue, and generation controls. MiniMax M2-her is developed by MiniMax. Adjust the available settings to shape the result for your task.
Model
M2-her
Documentation
MiniMax M2-her field reference
Use this reference to understand how every field in the playground changes the request sent to M2-her.
Stream response
Optional
Type: Boolean
Default: On
Sends small chunks as the model generates them, allowing the interface to display the beginning of a response sooner. It changes delivery timing rather than model quality; disable it when the client needs one complete response object.
Temperature
Optional
Type: Number
Default: 1
Range: 0.01–1
Adjusts how strongly the model favors its most likely next token. Lower values produce more focused and repeatable output, while higher values allow less-likely choices and increase variety; it only has an effect when sampling is enabled.
Top P
Optional
Type: Number
Default: 0.95
Range: 0–1
Limits sampling to the smallest set of likely tokens whose probabilities add up to this value. Lower settings make output more focused, while a value near 1 keeps more alternatives available; it only affects sampled generation.
Maximum output tokens
Optional
Type: Integer
Default: 2048
Range: 1–2048
Caps the total tokens used for hidden reasoning plus the final visible answer. Complex tasks may spend much of this budget reasoning, so a limit that is too low can truncate or weaken the final response.