Models · MiniMax

Deploy MiniMax models on dedicated GPUs

7 MiniMax models ready to deploy on GPU capacity reserved for you, from 220B to 456B parameters, covering chat & reasoning. Private OpenAI-compatible endpoint, European GPUs, billed per GPU-hour.

Request a deployment ↓MiniMaxAI on Hugging Face ↗All makersHow dedicated works
ModelParams
MiniMax-M3.1-FlashnewtrendingvisionMoE
MiniMax		220B
MiniMax-M3visionMoE
MiniMax		427B
MiniMax-M2.5MoE
MiniMax		229B
MiniMax-M2.7MoE
MiniMax		229B
MiniMax-M2MoE
MiniMax		229B
MiniMax-M2.1MoE
MiniMax		229B
MiniMax-M1-80kMoE
MiniMax		456B

Specs from Hugging Face, synced 2026-10-27. Machine-readable version: https://sference.com/models/MiniMaxAI.md

FAQ

Which MiniMax models can I deploy on sference?

7 today, most requested first: MiniMax-M3.1-Flash, MiniMax-M3, MiniMax-M2.5, MiniMax-M2.7, MiniMax-M2, MiniMax-M2.1 and MiniMax-M1-80k. Any other MiniMax checkpoint or a fine-tune of one can be requested with the form on this page.

How many GPUs do MiniMax models need?

It depends on the model: the lineup runs from 220B to 456B parameters. Each model page lists the checkpoint size and the minimum footprint on 80 GB and 141 GB GPUs. You don't pick hardware on sference: you choose a performance tier and we size the deployment.

Can I use MiniMax models commercially?

1 of the 7 listed models are under licenses that allow commercial use (Custom license, Modified MIT and Apache 2.0 across the lineup). Each model page summarizes its license and links the full text.

How is a dedicated deployment priced?

Per GPU-hour on a monthly commitment, sized to your throughput and latency targets.

Request a MiniMax model

Tell us which MiniMax model and roughly how much traffic you expect. Fine-tunes of these models work the same way.

  1. 01
    We size it

    A performance tier, the capacity behind it, and a quote for your volume.

  2. 02
    We benchmark it

    Bring one workload with a sample of real traffic; we come back with latency and throughput numbers.

  3. 03
    It goes live

    We provision the deployment. You call it with the same API and model id as before.

No commitment. Prefer email? [email protected]