Model Comparison

Qwen3.6-35B-A3B vs MAI-Thinking-1Which is better in 2026?

MAI-Thinking-1 shows notably better performance in the majority of benchmarks.

Verdict: Qwen3.6-35B-A3B vs MAI-Thinking-1 — which is better?

Qwen3.6-35B-A3B (by Alibaba Cloud / Qwen Team) and MAI-Thinking-1 (by Microsoft) are two of the AI models people compare most. Here is how they stack up on benchmarks, price and capabilities, and which one to pick in 2026.

Qwen3.6-35B-A3B outperforms in 3 benchmarks (GPQA, MMLU-Pro, Terminal-Bench 2.0), while MAI-Thinking-1 is better at 5 benchmarks (AIME 2026, HMMT Feb 26, LiveCodeBench v6, SWE-Bench Pro, SWE-Bench Verified). MAI-Thinking-1 shows notably better performance in the majority of benchmarks.

Choose Qwen3.6-35B-A3B if…

  • you need open weights you can self-host or fine-tune

Choose MAI-Thinking-1 if…

  • you want the strongest raw capability — it leads on 5 of 8 shared benchmarks
  • you want the most recent training data — it shipped Jun 2026

Performance Benchmarks

Comparative analysis across standard metrics

8 benchmarks

Qwen3.6-35B-A3B outperforms in 3 benchmarks (GPQA, MMLU-Pro, Terminal-Bench 2.0), while MAI-Thinking-1 is better at 5 benchmarks (AIME 2026, HMMT Feb 26, LiveCodeBench v6, SWE-Bench Pro, SWE-Bench Verified).

MAI-Thinking-1 shows notably better performance in the majority of benchmarks.

Fri Jul 17 2026 • llm-stats.com

Arena Performance

Human preference votes

Model Size

Parameter count comparison

965.0B diff

MAI-Thinking-1 has 965.0B more parameters than Qwen3.6-35B-A3B, making it 2757.1% larger.

Alibaba Cloud / Qwen Team
Qwen3.6-35B-A3B
35.0Bparameters
Microsoft
MAI-Thinking-1
1.0Tparameters
35.0B
Qwen3.6-35B-A3B
1000.0B
MAI-Thinking-1

Input Capabilities

Supported data types and modalities

Qwen3.6-35B-A3B supports multimodal inputs, whereas MAI-Thinking-1 does not.

Qwen3.6-35B-A3B can handle both text and other forms of data like images, making it suitable for multimodal applications.

Qwen3.6-35B-A3B

Text
Images
Audio
Video

MAI-Thinking-1

Text
Images
Audio
Video

License

Usage and distribution terms

Qwen3.6-35B-A3B 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.

Qwen3.6-35B-A3B

Apache 2.0

Open weights

MAI-Thinking-1

Proprietary

Closed source

Release Timeline

When each model was launched

Qwen3.6-35B-A3B was released on 2026-04-16, while MAI-Thinking-1 was released on 2026-06-02.

MAI-Thinking-1 is 2 months newer than Qwen3.6-35B-A3B.

Qwen3.6-35B-A3B

Apr 16, 2026

3 months ago

MAI-Thinking-1

Jun 2, 2026

1 months ago

1mo newer

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→

Key Takeaways

Alibaba Cloud / Qwen Team

Qwen3.6-35B-A3B

View details

Alibaba Cloud / Qwen Team

Supports multimodal inputs
Has open weights
Higher GPQA score (86.0% vs 84.2%)
Higher MMLU-Pro score (85.2% vs 85.0%)
Higher Terminal-Bench 2.0 score (51.5% vs 46.0%)
Microsoft
Higher AIME 2026 score (94.5% vs 92.7%)
Higher HMMT Feb 26 score (84.9% vs 83.6%)
Higher LiveCodeBench v6 score (87.7% vs 80.4%)
Higher SWE-Bench Pro score (52.8% vs 49.5%)
Higher SWE-Bench Verified score (73.5% vs 73.4%)

Detailed Comparison

Interactive Arena

Judge for yourself.

Run your own prompts against Qwen3.6-35B-A3B and MAI-Thinking-1 side-by-side, then vote on the output you prefer.

Qwen3.6-35B-A3B
✓ Preferred
MAI-Thinking-1
Open in Playground
AI Model Comparison Table
Feature
Alibaba Cloud / Qwen Team
Qwen3.6-35B-A3B
Microsoft
MAI-Thinking-1

FAQ

Common questions about Qwen3.6-35B-A3B vs MAI-Thinking-1.

Which is better, Qwen3.6-35B-A3B or MAI-Thinking-1?

MAI-Thinking-1 shows notably better performance in the majority of benchmarks. Qwen3.6-35B-A3B is made by Alibaba Cloud / Qwen Team and MAI-Thinking-1 is made by Microsoft. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Qwen3.6-35B-A3B compare to MAI-Thinking-1 in benchmarks?

Qwen3.6-35B-A3B scores MMLU-Redux: 93.3%, MMBench-V1.1: 92.8%, AI2D: 92.7%, AIME 2026: 92.7%, RefCOCO-avg: 92.0%. 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 main differences between Qwen3.6-35B-A3B and MAI-Thinking-1?

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

Who makes Qwen3.6-35B-A3B and MAI-Thinking-1?

Qwen3.6-35B-A3B is developed by Alibaba Cloud / Qwen Team and MAI-Thinking-1 is developed by Microsoft.

Related comparisons