The AI arena is free today

Open Superagent

Gemma 4 31B vs MAI-Thinking-1

Gemma 4 31B and MAI-Thinking-1 are closely matched at 33.1 and 33.0 on the LLM Stats Score.

Google · Microsoft · Updated for 2026

Which is better?

Gemma 4 31B and MAI-Thinking-1 are closely matched on the overall LLM Stats Score at 33.1 and 33.0.

In the 5 individual benchmarks reported for both models, Gemma 4 31B wins 3; 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 Gemma 4 31B

  • you value its reported benchmark strengths — it wins 3 of 5 exact shared results
  • you need open weights you can self-host or fine-tune

Choose MAI-Thinking-1

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

At a glance

The differences that matter most.

Gemma 4 31B
Google
MAI-Thinking-1
Microsoft
Core performance indexes
LLM Stats Score
33.1
#104
33.0
#105
Reasoning
33.6
#97
33.8
#96
Agents
13.6
#94
12.7
#101
Cost, coverage & limits
Benchmark wins
3 of 5
2 of 5
Input price
$0.09 / M
— / M
Output price
$0.34 / M
— / M
Context window
262,144
—

Capability indexes

Additional strengths measured across groups of related public benchmarks

2 shared
Index
Gemma 4 31B
MAI-Thinking-1
Math
29.4#84
33.7#53
Long context
16.7#45
20.4#30
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

12 reported for Gemma 4 31B · 23 for MAI-Thinking-1

5 shared

Gemma 4 31B outperforms in 3 benchmarks (GPQA, MedXpertQA, MMLU-Pro), while MAI-Thinking-1 is better at 2 benchmarks (AIME 2026, LiveCodeBench v6).

Gemma 4 31B has a slight edge in benchmark performance.

Tue Sep 22 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

969.3B diff

MAI-Thinking-1 has 969.3B more parameters than Gemma 4 31B, making it 3157.3% larger.

Google
Gemma 4 31B
30.7Bparameters
Microsoft
MAI-Thinking-1
1.0Tparameters
30.7B
Gemma 4 31B
1000.0B
MAI-Thinking-1

Context Window

Maximum input and output token capacity

Only Gemma 4 31B specifies input context (262,144 tokens). Only Gemma 4 31B specifies output context (262,144 tokens).

Google
Gemma 4 31B
Input262,144 tokens
Output262,144 tokens
Microsoft
MAI-Thinking-1
Input- tokens
Output- tokens
Tue Sep 22 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Gemma 4 31B supports multimodal inputs, whereas MAI-Thinking-1 does not.

Gemma 4 31B can handle both text and other forms of data like images, making it suitable for multimodal applications.

Gemma 4 31B

Text
Images
Audio
Video

MAI-Thinking-1

Text
Images
Audio
Video

License

Usage and distribution terms

Gemma 4 31B is licensed under Apache 2.0, while MAI-Thinking-1 uses a proprietary license.

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

Gemma 4 31B

Apache 2.0

Open weights

MAI-Thinking-1

Proprietary

Closed source

Release Timeline

When each model was launched

Gemma 4 31B was released on 2026-04-02, while MAI-Thinking-1 was released on 2026-06-02.

MAI-Thinking-1 is 2 months newer than Gemma 4 31B.

Gemma 4 31B

Apr 2, 2026

5 months ago

MAI-Thinking-1

Jun 2, 2026

3 months ago

2mo newer

Knowledge Cutoff

When training data ends

Gemma 4 31B has a documented knowledge cutoff of 2025-01-01, while MAI-Thinking-1's cutoff date is not specified.

We can confirm Gemma 4 31B's training data extends to 2025-01-01, but cannot make a direct comparison without MAI-Thinking-1's cutoff date.

Gemma 4 31B

Jan 2025

MAI-Thinking-1

—

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Judge for yourself.

Run your own prompts against Gemma 4 31B and MAI-Thinking-1 side-by-side, then vote on the output you prefer.

Gemma 4 31B
✓ Preferred
MAI-Thinking-1
Open in Playground

FAQ

Common questions about Gemma 4 31B vs MAI-Thinking-1.

Which is better, Gemma 4 31B or MAI-Thinking-1?

Gemma 4 31B and MAI-Thinking-1 are closely matched on the LLM Stats Score at 33.1 and 33.0. Gemma 4 31B is made by Google and MAI-Thinking-1 is made by Microsoft. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Gemma 4 31B compare to MAI-Thinking-1 in benchmarks?

Gemma 4 31B scores AIME 2026: 89.2%, MMMLU: 88.4%, t2-bench: 86.4%, MathVision: 85.6%, MMLU-Pro: 85.2%. MAI-Thinking-1 scores LongFact: 98.0%, AIME 2025: 97.0%, AIME 2026: 94.5%, GraphWalks: 90.0%, AIR-Bench: 88.0%.

What are the context window sizes for Gemma 4 31B and MAI-Thinking-1?

Gemma 4 31B supports 262K tokens and MAI-Thinking-1 supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Gemma 4 31B and MAI-Thinking-1?

Key differences include LLM Stats Score (33.1 vs 33.0), multimodal support (yes vs no), licensing (Apache 2.0 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes Gemma 4 31B and MAI-Thinking-1?

Gemma 4 31B is developed by Google and MAI-Thinking-1 is developed by Microsoft.

Related comparisons