MiniMax-M3
MiniMax · 1 configuration
Intelligence29.2
Input / 1M$0.30
Output / 1M$1.20
Context1M
Tariff: MiniMax · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost1 measured configurations · 0 without both values
Selected: thinking thinking · Click a point to switch configuration
MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
29.2 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
92.9 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
39.0 %SciCodeCode generation for scientific research problems.
47.1 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
65.2 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
88.9 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
82.9 %MMMU-ProMultimodal understanding of academic problems that include images.
78.6 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
16.7 %Specifications from the catalog
Inputtext, image, video
Outputtext
Tool callingYes
Structured outputYes
Speed107.5 tok/s
First token1.49 s
Maximum output512K
Catalog listing date2026-05-31