Model Comparison

North Mini Code 1.0 vs MAI-Thinking-1Which is better in 2026?

MAI-Thinking-1 significantly outperforms across most benchmarks.

Verdict: North Mini Code 1.0 vs MAI-Thinking-1 — which is better?

North Mini Code 1.0 (by Cohere) 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.

North Mini Code 1.0 outperforms in 0 benchmarks, while MAI-Thinking-1 is better at 4 benchmarks (LiveCodeBench v6, SWE-Bench Pro, SWE-Bench Verified, Terminal-Bench 2.0). MAI-Thinking-1 significantly outperforms across most benchmarks.

Choose North Mini Code 1.0 if…

  • you want the most recent training data — it shipped Jun 2026
  • 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 4 of 4 shared benchmarks

Performance Benchmarks

Comparative analysis across standard metrics

4 benchmarks

North Mini Code 1.0 outperforms in 0 benchmarks, while MAI-Thinking-1 is better at 4 benchmarks (LiveCodeBench v6, SWE-Bench Pro, SWE-Bench Verified, Terminal-Bench 2.0).

MAI-Thinking-1 significantly outperforms across most benchmarks.

Fri Jul 17 2026 • llm-stats.com

Arena Performance

Human preference votes

Model Size

Parameter count comparison

970.0B diff

MAI-Thinking-1 has 970.0B more parameters than North Mini Code 1.0, making it 3233.3% larger.

Cohere
North Mini Code 1.0
30.0Bparameters
Microsoft
MAI-Thinking-1
1.0Tparameters
30.0B
North Mini Code 1.0
1000.0B
MAI-Thinking-1

License

Usage and distribution terms

North Mini Code 1.0 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.

North Mini Code 1.0

Apache 2.0

Open weights

MAI-Thinking-1

Proprietary

Closed source

Release Timeline

When each model was launched

North Mini Code 1.0 was released on 2026-06-09, while MAI-Thinking-1 was released on 2026-06-02.

North Mini Code 1.0 is 0 month newer than MAI-Thinking-1.

North Mini Code 1.0

Jun 9, 2026

1 months ago

1w newer
MAI-Thinking-1

Jun 2, 2026

1 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

Key Takeaways

Has open weights
Higher LiveCodeBench v6 score (87.7% vs 70.3%)
Higher SWE-Bench Pro score (52.8% vs 40.2%)
Higher SWE-Bench Verified score (73.5% vs 67.6%)
Higher Terminal-Bench 2.0 score (46.0% vs 36.0%)

Detailed Comparison

Interactive Arena

Judge for yourself.

Run your own prompts against North Mini Code 1.0 and MAI-Thinking-1 side-by-side, then vote on the output you prefer.

North Mini Code 1.0
✓ Preferred
MAI-Thinking-1
Open in Playground
AI Model Comparison Table
Feature
Cohere
North Mini Code 1.0
Microsoft
MAI-Thinking-1

FAQ

Common questions about North Mini Code 1.0 vs MAI-Thinking-1.

Which is better, North Mini Code 1.0 or MAI-Thinking-1?

MAI-Thinking-1 significantly outperforms across most benchmarks. North Mini Code 1.0 is made by Cohere 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 North Mini Code 1.0 compare to MAI-Thinking-1 in benchmarks?

North Mini Code 1.0 scores LiveCodeBench v6: 70.3%, SWE-Bench Verified: 67.6%, SWE-Bench Pro: 40.2%, SciCode: 38.2%, Terminal-Bench 2.0: 36.0%. 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 North Mini Code 1.0 and MAI-Thinking-1?

Key differences include licensing (Apache 2.0 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes North Mini Code 1.0 and MAI-Thinking-1?

North Mini Code 1.0 is developed by Cohere and MAI-Thinking-1 is developed by Microsoft.