Model Comparison

Codestral-22B vs Gemma 3n E4B InstructedWhich is better in 2026?

Codestral-22B significantly outperforms across most benchmarks.

Verdict: Codestral-22B vs Gemma 3n E4B Instructed — which is better?

Codestral-22B (by Mistral AI) and Gemma 3n E4B Instructed (by Google) 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.

Codestral-22B outperforms in 2 benchmarks (HumanEval, MBPP), while Gemma 3n E4B Instructed is better at 0 benchmarks. Codestral-22B significantly outperforms across most benchmarks.

Choose Codestral-22B if…

  • you want the strongest raw capability — it leads on 2 of 2 shared benchmarks
  • you need open weights you can self-host or fine-tune

Choose Gemma 3n E4B Instructed if…

  • you want the most recent training data — it shipped Jun 2025
Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

2 benchmarks

Codestral-22B outperforms in 2 benchmarks (HumanEval, MBPP), while Gemma 3n E4B Instructed is better at 0 benchmarks.

Codestral-22B significantly outperforms across most benchmarks.

Fri Jun 12 2026 • llm-stats.com

Arena Performance

Human preference votes

Model Size

Parameter count comparison

14.2B diff

Codestral-22B has 14.2B more parameters than Gemma 3n E4B Instructed, making it 177.5% larger.

Mistral AI
Codestral-22B
22.2Bparameters
Google
Gemma 3n E4B Instructed
8.0Bparameters
22.2B
Codestral-22B
8.0B
Gemma 3n E4B Instructed

Context Window

Maximum input and output token capacity

Only Gemma 3n E4B Instructed specifies input context (32,000 tokens). Only Gemma 3n E4B Instructed specifies output context (32,000 tokens).

Mistral AI
Codestral-22B
Input- tokens
Output- tokens
Google
Gemma 3n E4B Instructed
Input32,000 tokens
Output32,000 tokens
Fri Jun 12 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Gemma 3n E4B Instructed supports multimodal inputs, whereas Codestral-22B does not.

Gemma 3n E4B Instructed can handle both text and other forms of data like images, making it suitable for multimodal applications.

Codestral-22B

Text
Images
Audio
Video

Gemma 3n E4B Instructed

Text
Images
Audio
Video

License

Usage and distribution terms

Codestral-22B is licensed under MNPL-0.1, while Gemma 3n E4B Instructed uses a proprietary license.

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

Codestral-22B

MNPL-0.1

Open weights

Gemma 3n E4B Instructed

Proprietary

Closed source

Release Timeline

When each model was launched

Codestral-22B was released on 2024-05-29, while Gemma 3n E4B Instructed was released on 2025-06-26.

Gemma 3n E4B Instructed is 13 months newer than Codestral-22B.

Codestral-22B

May 29, 2024

2.0 years ago

Gemma 3n E4B Instructed

Jun 26, 2025

11 months ago

1.1yr newer

Knowledge Cutoff

When training data ends

Gemma 3n E4B Instructed has a documented knowledge cutoff of 2024-06-01, while Codestral-22B's cutoff date is not specified.

We can confirm Gemma 3n E4B Instructed's training data extends to 2024-06-01, but cannot make a direct comparison without Codestral-22B's cutoff date.

Codestral-22B

—

Gemma 3n E4B Instructed

Jun 2024

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

Mistral AI
Has open weights
Higher HumanEval score (81.1% vs 75.0%)
Higher MBPP score (78.2% vs 63.6%)
Google
Larger context window (32,000 tokens)
Supports multimodal inputs

Detailed Comparison

AI Model Comparison Table
Feature
Mistral AI
Codestral-22B
Google
Gemma 3n E4B Instructed

FAQ

Common questions about Codestral-22B vs Gemma 3n E4B Instructed.

Which is better, Codestral-22B or Gemma 3n E4B Instructed?

Codestral-22B significantly outperforms across most benchmarks. Codestral-22B is made by Mistral AI and Gemma 3n E4B Instructed is made by Google. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Codestral-22B compare to Gemma 3n E4B Instructed in benchmarks?

Codestral-22B scores HumanEvalFIM-Average: 91.6%, HumanEval: 81.1%, MBPP: 78.2%, Spider: 63.5%, HumanEval-Average: 61.5%. Gemma 3n E4B Instructed scores HumanEval: 75.0%, MGSM: 67.0%, MMLU: 64.9%, Global-MMLU-Lite: 64.5%, MBPP: 63.6%.

What are the context window sizes for Codestral-22B and Gemma 3n E4B Instructed?

Codestral-22B supports an unknown number of tokens and Gemma 3n E4B Instructed supports 32K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Codestral-22B and Gemma 3n E4B Instructed?

Key differences include multimodal support (no vs yes), licensing (MNPL-0.1 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes Codestral-22B and Gemma 3n E4B Instructed?

Codestral-22B is developed by Mistral AI and Gemma 3n E4B Instructed is developed by Google.

Related comparisons