The AI arena is free today

Open Superagent

Codestral-22B vs Gemini Diffusion

Codestral-22B and Gemini Diffusion are closely matched at -0.5 and 0.0 on the LLM Stats Score.

Mistral AI · Google · Updated for 2026

Which is better?

Codestral-22B and Gemini Diffusion are closely matched on the overall LLM Stats Score at -0.5 and 0.0.

The models split the 2 individual benchmarks reported for both models evenly.

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

Choose Codestral-22B

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

Choose Gemini Diffusion

  • you want the most recent training data — it shipped May 2025

At a glance

The differences that matter most.

Core performance indexes
-0.5
#381
0.0
#378
-0.6
#372
0.1
#369
2.3
#275
2.4
#274
Cost, coverage & limits
Benchmark wins
1 of 2
1 of 2
Input price
— / M
— / M
Output price
— / M
— / M
Context window
—
—

Individual benchmarks

7 reported for Codestral-22B · 10 for Gemini Diffusion

2 shared

Codestral-22B outperforms in 1 benchmarks (MBPP), while Gemini Diffusion is better at 1 benchmark (HumanEval).

Both models are evenly matched across the benchmarks.

Sun Nov 22 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

License

Usage and distribution terms

Codestral-22B is licensed under MNPL-0.1, while Gemini Diffusion 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

Gemini Diffusion

Proprietary

Closed source

Release Timeline

When each model was launched

Codestral-22B was released on 2024-05-29, while Gemini Diffusion was released on 2025-05-20.

Gemini Diffusion is 12 months newer than Codestral-22B.

Codestral-22B

May 29, 2024

2.5 years ago

Gemini Diffusion

May 20, 2025

1.5 years ago

11mo 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?

Judge for yourself.

Run your own prompts against Codestral-22B and Gemini Diffusion side-by-side, then vote on the output you prefer.

Codestral-22B
✓ Preferred
Gemini Diffusion
Open in Playground

FAQ

Common questions about Codestral-22B vs Gemini Diffusion.

Which is better, Codestral-22B or Gemini Diffusion?

Codestral-22B and Gemini Diffusion are closely matched on the LLM Stats Score at -0.5 and 0.0. Codestral-22B is made by Mistral AI and Gemini Diffusion is made by Google. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Codestral-22B compare to Gemini Diffusion in benchmarks?

Codestral-22B scores HumanEvalFIM-Average: 91.6%, HumanEval: 81.1%, MBPP: 78.2%, Spider: 63.5%, HumanEval-Average: 61.5%. Gemini Diffusion scores HumanEval: 89.6%, MBPP: 76.0%, Global-MMLU-Lite: 69.1%, LBPP (v2): 56.8%, BigCodeBench: 45.4%.

What are the main differences between Codestral-22B and Gemini Diffusion?

Key differences include LLM Stats Score (-0.5 vs 0.0), licensing (MNPL-0.1 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes Codestral-22B and Gemini Diffusion?

Codestral-22B is developed by Mistral AI and Gemini Diffusion is developed by Google.