The AI arena is free today

Open Superagent

MAI-Thinking-1 vs MiniMax M3

MiniMax M3 leads the LLM Stats Score 41.4 to 33.0.

Microsoft · MiniMax · Updated for 2026

Which is better?

MiniMax M3 leads the overall LLM Stats Score 41.4 to 33.0, ranking #50 overall.

In the 2 individual benchmarks reported for both models, MiniMax M3 wins 2; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose MAI-Thinking-1

  • you want the most recent training data — it shipped Jun 2026

Choose MiniMax M3

  • overall performance matters — it scores 41.4 and ranks #50 on LLM Stats
  • your work emphasizes reasoning and coding — it leads those capability indexes
  • you value its reported benchmark strengths — it wins 2 of 2 exact shared results
  • you need open weights you can self-host or fine-tune

At a glance

The differences that matter most.

Core performance indexes
33.0
#101
41.4
#50
33.8
#92
41.4
#48
19.4
#104
32.1
#39
12.8
#97
26.5
#48
Cost, coverage & limits
Benchmark wins
0 of 2
2 of 2
Input price
— / M
$0.28 / M
Output price
— / M
$1.10 / M
Context window
524,288

Capability indexes

Additional strengths measured across groups of related public benchmarks

2 shared
Index
MAI-Thinking-1
MiniMax M3
33.7#53
25.6#107
10.7#119
21.7#49
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

23 reported for MAI-Thinking-1 · 35 for MiniMax M3

2 shared

MAI-Thinking-1 outperforms in 0 benchmarks, while MiniMax M3 is better at 2 benchmarks (SWE-Bench Pro, SWE-Bench Verified).

MiniMax M3 significantly outperforms across most benchmarks.

Mon Sep 14 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

572.0B diff

MAI-Thinking-1 has 572.0B more parameters than MiniMax M3, making it 133.6% larger.

Microsoft
MAI-Thinking-1
1.0Tparameters
MiniMax
MiniMax M3
428.0Bparameters
1000.0B
MAI-Thinking-1
428.0B
MiniMax M3

Context Window

Maximum input and output token capacity

Only MiniMax M3 specifies input context (524,288 tokens). Only MiniMax M3 specifies output context (524,288 tokens).

Microsoft
MAI-Thinking-1
Input- tokens
Output- tokens
MiniMax
MiniMax M3
Input524,288 tokens
Output524,288 tokens
Mon Sep 14 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

MiniMax M3 supports multimodal inputs, whereas MAI-Thinking-1 does not.

MiniMax M3 can handle both text and other forms of data like images, making it suitable for multimodal applications.

MAI-Thinking-1

Text
Images
Audio
Video

MiniMax M3

Text
Images
Audio
Video

License

Usage and distribution terms

MAI-Thinking-1 is licensed under a proprietary license, while MiniMax M3 uses MIT.

License differences may affect how you can use these models in commercial or open-source projects.

MAI-Thinking-1

Proprietary

Closed source

MiniMax M3

MIT

Open weights

Release Timeline

When each model was launched

MAI-Thinking-1 was released on 2026-06-02, while MiniMax M3 was released on 2026-06-01.

MAI-Thinking-1 is 0 month newer than MiniMax M3.

MAI-Thinking-1

Jun 2, 2026

3 months ago

1d newer
MiniMax M3

Jun 1, 2026

3 months ago

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against MAI-Thinking-1 and MiniMax M3 side-by-side, then vote on the output you prefer.

MAI-Thinking-1
✓ Preferred
MiniMax M3
Open in Playground

FAQ

Common questions about MAI-Thinking-1 vs MiniMax M3.

Which is better, MAI-Thinking-1 or MiniMax M3?

MiniMax M3 leads the LLM Stats Score 41.4 to 33.0. MAI-Thinking-1 is made by Microsoft and MiniMax M3 is made by MiniMax. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does MAI-Thinking-1 compare to MiniMax M3 in benchmarks?

MAI-Thinking-1 scores LongFact: 98.0%, AIME 2025: 97.0%, AIME 2026: 94.5%, GraphWalks: 90.0%, AIR-Bench: 88.0%. MiniMax M3 scores OmniDocBench 1.5: 91.6%, SpreadSheetBench-v1: 89.3%, USAMO 2026: 85.7%, Video-MME: 85.4%, VideoMMMU: 84.6%.

What are the context window sizes for MAI-Thinking-1 and MiniMax M3?

MAI-Thinking-1 supports an unknown number of tokens and MiniMax M3 supports 524K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between MAI-Thinking-1 and MiniMax M3?

Key differences include LLM Stats Score (33.0 vs 41.4), multimodal support (no vs yes), licensing (Proprietary vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes MAI-Thinking-1 and MiniMax M3?

MAI-Thinking-1 is developed by Microsoft and MiniMax M3 is developed by MiniMax.