TokenTally

MiniMax Open Platform

MiniMax M2 cost planner

Use the 204K token window and low token rates to stretch batch coding, automation, and multi-step agent runs.

Last pricing check: Mar 13, 2026

$0.30 per 1M prompt tokens$1.20 per 1M completion tokens204,800 token context

Why teams choose this model

General-purpose coding copilots
Agentic tool runners
Enterprise chat copilots

Scenario planning

Realistic cost examples

Numbers use MiniMax M2 pricing

Repo refactor bot

Developers kick off automated refactors across mono-repo services.

Per request

$0.0018

Per month

$16.20

Tokens sent

24,300,000

1600 prompt tokens1100 completion tokens9,000 requests/mo

Ops agent

On-call copilots summarize incidents and execute shell actions with tool use.

Per request

$0.0011

Per month

$19.98

Tokens sent

28,800,000

900 prompt tokens700 completion tokens18,000 requests/mo

Multilingual assistant

Internal chat bot handles policy questions across languages on Bedrock standard tier.

Per request

$0.0007

Per month

$30.78

Tokens sent

45,900,000

600 prompt tokens420 completion tokens45,000 requests/mo

Compare with

FAQs

Where do these prices come from?

We reference the AWS Bedrock standard tier in US East/West ($0.30/M input, $1.20/M output). Flex/priority tiers shift that up or down if you need them.

How large is the context window?

MiniMax’s API docs list a 204,800-token window for both M2 and M2.1, which TokenTally now surfaces in the UI.

Does M2 support tool calling?

Yes—MiniMax exposes function calling + agent orchestration. Our scenario costs assume standard tool usage without extra provider surcharges.

Pricing sources