Polarison

MiniMax · open weights

MiniMax M3 API pricing by provider

MiniMax M3 is available from 11 providers. The cheapest is DeepInfra at $0.224 input / $0.88 output per 1M tokens (FP8). MiniMax’s own API charges $0.30 / $1.20, 26% more than the cheapest host. The priciest host, Mara, charges 2.7x the cheapest.

Prices from the OpenRouter API, retrieved October 26, 2026. Weights: MiniMaxAI/Minimax-M3.

ProviderPrecisionInput / 1MOutput / 1MDocument Q&A / monthContextMax output
DeepInfrafp8$0.224$0.88$116.00524K131K
CoreWeavefp4$0.23$0.96$120.80262K236K
GMICloudfp8$0.24$0.96$124.801.05M524K
StreamLakefp8$0.30$1.20$156.001M512K
Together—$0.30$1.20$156.00524K472K
Parasailfp8$0.30$1.20$156.001.05M524K
AtlasCloudfp8$0.30$1.20$156.00524K524K
Novitafp8$0.30$1.20$156.001M131K
Minimaxcreatorfp8$0.30$1.20$156.00524K512K
ModelRunfp8$0.30$1.30$159.001.05M944K
Mara—$0.60$2.40$312.001.05M944K

Sorted by blended price (3 input : 1 output). Document Q&A assumes 50,000 questions a month with 8,000 input and 600 output tokens each. Precision is reported by the provider; “—” means not disclosed.

Frequently asked questions

What is the cheapest MiniMax M3 API provider?
DeepInfra is the cheapest healthy provider we track at $0.224 per 1M input tokens and $0.88 per 1M output tokens.
How much does MiniMax M3 cost?
Across 11 providers, input prices range from $0.224 to $0.60 per 1M tokens and output prices from $0.88 to $2.40.
What is MiniMax M3’s context window?
MiniMax M3 supports up to 1.05M tokens, but some providers serve a smaller context window — check the table before choosing.

Other open-weight models