MiniMax Integration

MiniMax is a Chinese AI company offering high-performance LLMs via an OpenAI-compatible API. Embabel integrates MiniMax as a first-class provider using the same OpenAiCompatibleModelFactory pattern as other OpenAI-compatible providers.

Add the Dependency

<dependency>
    <groupId>com.embabel.agent</groupId>
    <artifactId>embabel-agent-starter-minimax</artifactId>
</dependency>

API Key Configuration

Set your MiniMax API key via environment variable (recommended) or Spring property:

export MINIMAX_API_KEY=your-api-key

Or in application.yml:

embabel:
  agent:
    platform:
      models:
        minimax:
          api-key: your-api-key

Use the property for local development and the environment variable in production deployments.

Available Models

MiniMax-M2.7 is the flagship model, optimized for quality. MiniMax-M2.7-highspeed trades some quality for significantly lower latency and cost — a good choice for intermediate steps in a multi-action agent flow.

Using MiniMax Models

Reference models by name in @LlmCall or programmatically via ai.withLlm():

// Declarative
@LlmCall(llm = "MiniMax-M2.7")
fun summarize(article: Article): Summary

// Programmatic
ai.withLlm("MiniMax-M2.7-highspeed")
    .create<Classification>("Classify this input")

Or map MiniMax models to roles in your configuration:

embabel:
  models:
    llms:
      cheapest: MiniMax-M2.7-highspeed
      best: MiniMax-M2.7

Then reference by role with the # prefix:

@LlmCall(llm = "#cheapest")
fun extractEntities(text: String): EntityList

Temperature Clamping

MiniMax models require temperature in the range (0.0, 1.0] — a value of exactly 0.0 is not permitted. Embabel’s MiniMaxOptionsConverter clamps temperature automatically:

• Values <= 0.0 are raised to 0.01
• Values > 1.0 are lowered to 1.0

A DEBUG log message is emitted whenever clamping occurs. No action is required on your part — this is handled transparently.

Configuration Reference

embabel:
  agent:
    platform:
      models:
        minimax:
          api-key: your-api-key               # Alternative to MINIMAX_API_KEY env var
          base-url: https://api.minimax.io/v1  # Default; override for proxies
          max-attempts: 4                      # Retry attempts (default: 4)
          backoff-millis: 1500                 # Initial backoff ms (default: 1500)
          backoff-multiplier: 2.0              # Backoff multiplier (default: 2.0)
          backoff-max-interval: 60000          # Max backoff ms (default: 60000)

See Also

• Working with LLMs
• Configuration Reference
• MiniMax AI