All models

MiniMax M3.1 Flash

MiniMax1M contextVisionReasoningTool calling

As of Oct 19, 2026, MiniMax M3.1 Flash costs $0.15 per million input tokens and $0.60 per million output tokens from MiniMax. No price change since Appndo started tracking it on Oct 19, 2026.

Official price from MiniMax

Pricing page

Per million tokens

Input
$0.15
Output
$0.60
Cached input
$0.015
Batch input / output
—
  • Up to 512K output tokens per request.

Estimate your monthly cost

Est. per month = input tokens × input price + output tokens × output price. Prices are per million tokens, so 10M input tokens at $3.00 cost $30. ¹ The price is higher for long prompts (usually over 200K tokens); the estimate uses the base price.

From MiniMax: $2.70 a month.

Where to buy it (1)

Sorted by your monthly estimate. Prices per million tokens.

  • MiniMaxOfficial

    Input / output
    $0.15 / $0.60
    Context
    1M
    Per month
    $2.70

    MiniMax-M3.1-Flash

Cloud and hosted sellers may add regional pricing, minimums or platform fees. Supports: Vision, Reasoning, Tool calling.

Similar models

About this data

Prices come from LiteLLM's model cost map, with more sellers from models.dev (both MIT licence), checked Oct 19, 2026; confirm on the seller's pricing page before you rely on a number.