Skip to content

MiniMax M3

minimax/minimax-m3
Language

The newer MiniMax at the same price, with a 524K context.

Model overview

Context window

524K

tokens

Input price

R6.20

per 1M tokens

Output price

R24.90

per 1M tokens

Weekly volume

tokens / week

Open weights

No

API only

Capabilities

3

of 11 tags

Description

MiniMax M3 is a language model from MiniMax. Webparam does not stock this model yet and can source it on request — tell us what you need it for and we will come back with availability and a price. MiniMax’s stated focus is multimodal breadth: text, video, speech. The weights are not published, so it is API-only.

Capabilities

Tool calling
Calls functions you define and returns their arguments as structured data, the basis of agents.
JSON mode
Constrains the response to valid JSON matching a schema you supply.
Streaming
Emits tokens as they are generated, so answers appear progressively rather than all at once.

Strengths & weaknesses

Strengths

  • Same price as M2.7 with more than twice the context
  • 524K context
  • Price quoted from the supplier's own rate card, not estimated

Weaknesses

  • Not provisioned on this account yet — lead time applies

Pricing

Pricing for MiniMax M3, per 1M tokens
RatePriceUnit
InputR6.20per 1M tokens
OutputR24.90per 1M tokens

This model is not provisioned yet; the rate is quoted on request.

Context window

524Ktokens

covers prompt and response together, so a long input leaves less room for the answer.

524K tokens against a peer range of 32K to 10M, median 256K. Compared across 34 language models. Logarithmic scale.

Supported features

Feature support for MiniMax M3
FeatureSupport
Tool callingSupported
JSON modeSupported
StreamingSupported
Vision inputNot supported
Audio inputNot supported
Long contextNot supported
Open weightsNot supported
ReasoningNot supported
Fine-tunableNot supported
BatchNot supported
CachingNot supported

Example use cases

  • General assistants

  • Retrieval-augmented answering

  • Everyday production traffic

MiniMax M3 compared with MiniMax M2.7 and AC Flash 4.5
AttributeMiniMax M3MiniMax M2.7AC Flash 4.5
Context window524K197K
Input priceR6.20R6.20R10.00
Output priceR24.90R24.90R40.00
Tool callingSupportedSupportedSupported
JSON modeSupportedSupportedSupported
Vision inputNot supportedNot supportedNot supported
Open weightsNot supportedNot supportedNot supported
Prompt cachingNot supportedNot supportedNot supported

Best value in each row is highlighted. Blank cells are facts the supplier does not publish, not zeroes.

Provider information

View MiniMax
MiniMax6 models

MiniMax · 稀宇科技 · Shanghai, China · est. 2021

MiniMax builds across more modalities than almost any peer, million-token text models, the Hailuo video line, and production-grade speech synthesis. Its M-series brought efficient open reasoning to the mix, rounding out one of the most complete catalogues in the ecosystem.

Recent releases

    Documentation

    Frequently asked questions

    Input is billed at R6.20 per 1M tokens and output at R24.90 per 1M tokens. Every figure on this page is an illustrative placeholder.