Skip to content

MiniMax

MiniMax M3

M3 can reason over up to one million tokens in one request, enough to compare lengthy specifications or trace dependencies across a large codebase.

Model interface
API model IDminimax/minimax-m3
Input
Chat messages
Output
Text
Endpoint
POST /v1/chat/completions

Best for

Technical details

Read the model documentation
API model ID
minimax/minimax-m3
Provider
MiniMax
Modality
Text
Input
Chat messages
Output
Text
Endpoint
POST /v1/chat/completions

Request parameters

  • messagesobject[] · Required

    The conversation, in OpenAI's format.

    OpenAI chat messages (system, user, assistant, tool)

  • max_completion_tokensinteger · Optional · Default: the model's

    Thinking tokens count toward it: too small a cap ends with finish_reason: length and an empty answer.

    1 up to the model's output limit

  • streamboolean · Optional · Default: false

    Server-sent events, as in OpenAI's API.

    • true
    • false
  • temperaturenumber · Optional · Default: 1

    MiniMax documents it as having no effect on M3 (fixed at 1).

    (0, 1]

  • toolsobject[] · Optional

    Function calling, in OpenAI's format.

    OpenAI function tools

Additional parameters

  • max_tokensinteger · Optional

    The older name for the same cap; MiniMax marks it deprecated.

    as above

  • reasoning_splitboolean · Optional · Default: the vendor's

    true returns the reasoning in reasoning_content; false leaves it inside the answer between think tags.

    • true
    • false
  • service_tierstring · Optional · Default: standard

    MiniMax's priority tier, on M3 only, billed at 1.5× the standard rates.

    priority

  • stream_optionsobject · Optional

    Adds a final chunk with token usage.

    {"include_usage": true}

  • thinkingobject · Optional · Default: on

    The model reasons before answering by default; disabled turns it off.

    {"type": "adaptive"} · {"type": "disabled"}

  • tool_choicestring · Optional · Default: auto

    Whether the model may, must, or must not call a tool, as MiniMax documents for M3.

    • none
    • auto
    • required
  • top_pnumber · Optional · Default: 0.95

    MiniMax documents it as having no effect on M3 (fixed at 0.95).

    (0, 1]

Pricing

Billable unit and optionsRate (USD)
Per 1M input tokens
Base rate$0.36
prompts over 512,000 tokens$0.72
priority (service_tier: priority)$0.54
Per 1M output tokens
Base rate$1.44
prompts over 512,000 tokens$2.88
priority (service_tier: priority)$2.16
Per 1M cached input tokens
Base rate$0.072
prompts over 512,000 tokens$0.144
priority (service_tier: priority)$0.108
Per 1M cache-write tokens
Base ratefree$0.00
priority (service_tier: priority)free$0.00

Billing notes

  • A prompt over 512,000 tokens bills the whole request at the long-context rates.

Catalogue rates on 21 Nov 2026. Prices exclude VAT.

View in pricing catalogue