curl --request POST \
  --url https://direct.evolink.ai/v1/messages \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "MiniMax-M3.1-Flash",
  "max_tokens": 131072,
  "messages": [
    {
      "role": "user",
      "content": "Hello! How can you help me?"
    }
  ]
}
'

{
  "id": "072514f09a86ce2abb0c5c3930f924fb",
  "type": "message",
  "role": "assistant",
  "model": "MiniMax-M3.1-Flash",
  "content": [
    {
      "thinking": "Let the ball cost x. Then 2x + 1 = 1.10, so x = 0.05.",
      "signature": "84eddf556420e0cc237f8f50212216543e8add2ddd931c9a271d2a5a95689333",
      "type": "thinking"
    },
    {
      "text": "0.05",
      "type": "text"
    }
  ],
  "usage": {
    "input_tokens": 55,
    "output_tokens": 66,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 174,
    "output_tokens_details": {
      "thinking_tokens": 63
    }
  },
  "stop_reason": "end_turn",
  "base_resp": {
    "status_code": 0,
    "status_msg": ""
  }
}

Messages API

MiniMax-M3.1-Flash - Anthropic-Compatible API

Call MiniMax-M3.1-Flash using the Anthropic Messages protocol
Request and response structures follow the Anthropic API
Multimodal conversations: text, image, and video blocks in content
System prompts: use the top-level system field
Reasoning: always enabled; adjust depth with output_config.effort. Reasoning is returned in content[type=thinking] blocks
Streaming: SSE events
Tool calling: Anthropic tool_use / tool_result flow

Streaming events: message_start → content_block_start → content_block_delta (thinking_delta / signature_delta / text_delta / input_json_delta) → content_block_stop → message_delta → message_stop. ping events may appear between these events. Use the usage in the message_delta event for final token counts.

POST

messages

curl --request POST \
  --url https://direct.evolink.ai/v1/messages \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "MiniMax-M3.1-Flash",
  "max_tokens": 131072,
  "messages": [
    {
      "role": "user",
      "content": "Hello! How can you help me?"
    }
  ]
}
'

{
  "id": "072514f09a86ce2abb0c5c3930f924fb",
  "type": "message",
  "role": "assistant",
  "model": "MiniMax-M3.1-Flash",
  "content": [
    {
      "thinking": "Let the ball cost x. Then 2x + 1 = 1.10, so x = 0.05.",
      "signature": "84eddf556420e0cc237f8f50212216543e8add2ddd931c9a271d2a5a95689333",
      "type": "thinking"
    },
    {
      "text": "0.05",
      "type": "text"
    }
  ],
  "usage": {
    "input_tokens": 55,
    "output_tokens": 66,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 174,
    "output_tokens_details": {
      "thinking_tokens": 63
    }
  },
  "stop_reason": "end_turn",
  "base_resp": {
    "status_code": 0,
    "status_msg": ""
  }
}

BaseURL: The default BaseURL is https://direct.evolink.ai, which has better support for text models and long-lived connections. https://api.evolink.ai is the primary endpoint for multimodal services and serves as a fallback address for text models.

Authorizations

Authorization

string

header

required

All APIs require Bearer Token authentication

Get API Key:

Visit API Key Management Page to get your API Key

Add to request header:

Authorization: Bearer YOUR_API_KEY

Note: EvoLink uses Bearer Token authentication uniformly for /v1/messages.

Body

application/json

model

enum<string>

required

Model to call

Available options:

MiniMax-M3.1-Flash

Example:

"MiniMax-M3.1-Flash"

messages

object[]

required

List of conversation messages, alternating user / assistant turns

Notes:

Must contain at least 1 message
The last message is typically role=user

Minimum array length: 1

Show child attributes

max_tokens

integer

Maximum number of generated tokens, including reasoning and the final answer.

Recommended: 131,072 (128K); maximum: 524,288 (512K)
A small budget may be exhausted during reasoning, resulting in stop_reason=max_tokens without text content

Required range: 1 <= x <= 524288

Example:

131072

system

System prompt used to set the assistant's role and behavior. Accepts a string or an array of text content blocks in the top-level system field.

Example:

"You are a helpful assistant."

temperature

number

default:1

Sampling temperature.

Range: [0, 2]; default: 1
Higher values produce more varied output; lower values produce more deterministic output
Adjust either temperature or top_p, rather than both

Required range: 0 <= x <= 2

Example:

1

top_p

number

default:0.95

Nucleus sampling threshold.

Range: [0, 1]; default: 0.95
Adjust either temperature or top_p, rather than both

Required range: 0 <= x <= 1

Example:

0.95

stream

boolean

default:false

Whether to return via SSE streaming

true: Server-Sent Events streaming response
false: Wait for complete response before returning (default)

Example:

false

thinking

object

Reasoning mode. MiniMax-M3.1-Flash always uses reasoning. Only adaptive is supported; this field may be omitted. Reasoning is returned in type="thinking" blocks and billed as output tokens.

Show child attributes

output_config

object

Output configuration.

Show child attributes

tools

object[]

Tool definition list

Notes:

Follows the Anthropic tool definition specification
input_schema uses a JSON Schema object

Show child attributes

tool_choice

object

Tool selection strategy. Only auto is supported; the model decides whether to call a tool.

Show child attributes

metadata

object

Request metadata

Show child attributes

Response

Message object

Anthropic-style message response

string

Unique message ID

type

enum<string>

Response object type

Available options:

message

role

enum<string>

Available options:

assistant

model

string

Model actually used

Example:

"MiniMax-M3.1-Flash"

content

object[]

Response content blocks.

Possible types:

thinking: reasoning content, which may be absent for simple questions
text: final answer
tool_use: a tool call initiated by the model

Show child attributes

stop_reason

enum<string>

Stop reason

end_turn: Natural completion
max_tokens: Reached max_tokens limit
tool_use: Model triggered a tool call

Available options:

end_turn,

max_tokens,

tool_use

usage

object

Token usage statistics (Anthropic specification)

Show child attributes

base_resp

object

Status code and error details

Show child attributes

Image Series

Video Series

Audio Series

Text Series

Account Management

Task Management

File Management

MiniMax-M3.1-Flash - Anthropic-Compatible API

Authorizations

Body

Response