Frontier models
minimaxAGNT

MiniMax in AGNT

Efficient frontier models with huge context. Every AGNT workflow is provider-portable, so the same agents run here or anywhere else you point them.

Why teams pick it

Models

  • MiniMax-M2

Strengths

  • million-token context
  • strong coding
  • cost-efficient serving

In AGNT

Agents, workflows and tools use MiniMax like any other provider — same nodes, same receipts, swap models per step if you want.

Very large context, efficiently served

MiniMax combines an extremely large context window with cost-efficient serving, which is an unusual pairing — large windows are normally expensive to use. For workloads that genuinely need to hold an entire codebase or document set in view, that combination is the point.

What to watch

A million-token window is a capability, not an instruction. Filling it still costs money and dilutes attention, so retrieval discipline continues to pay even when the limit is no longer binding. Verify coding performance against your own repository rather than a generic benchmark.

Using MiniMax in a workflow

The practical question is not whether MiniMax is good but which steps deserve it. Point MiniMax-M2 at the judgment calls — the places where million-token context changes the answer — and route the high-volume routine work elsewhere in the same workflow. Swapping either side later does not touch the workflow itself.

Connect MiniMax in two minutes

  1. Download AGNT Community Core — free, local-first, no account needed to run.
  2. Open Settings → Providers, choose MiniMax, and paste your API key. Connect by pasting an API key into AGNT’s vault — stored encrypted on your machine, never uploaded.
  3. Give an agent a job — or install one from the marketplace — and watch the first receipt come back.

MiniMax in AGNT — common questions

What is a million-token context actually for?

Whole-repository reasoning, large document sets and long agent histories that would otherwise need aggressive summarisation.

Does a big window remove the need for RAG?

No — selecting the right material is still cheaper, faster and more accurate than sending everything.

Is MiniMax good for code?

Coding is a stated strength; evaluate it on your own codebase before committing a workflow to it.

Give AI a job. Get the proof.