The AI arena is free today

Open Superagent

Codestral-22B vs Qwen2.5-Coder 32B Instruct

Codestral-22B and Qwen2.5-Coder 32B Instruct are closely matched at 0.5 and 2.7 on the LLM Stats Score.

Mistral AI · Alibaba Cloud / Qwen Team · Updated for 2026

Which is better?

Codestral-22B and Qwen2.5-Coder 32B Instruct are closely matched on the overall LLM Stats Score at 0.5 and 2.7.

In the 2 individual benchmarks reported for both models, Qwen2.5-Coder 32B Instruct wins 2; 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 Codestral-22B

  • you are already invested in the Mistral AI ecosystem

Choose Qwen2.5-Coder 32B Instruct

  • you value its reported benchmark strengths — it wins 2 of 2 exact shared results
  • you want the most recent training data — it shipped Sep 2024

At a glance

The differences that matter most.

Core performance indexes
0.5
#304
2.7
#295
0.4
#297
2.6
#285
2.6
#209
10.1
#151
Cost, coverage & limits
Benchmark wins
0 of 2
2 of 2
Input price
— / M
$0.09 / M
Output price
— / M
$0.09 / M
Context window
128,000

Individual benchmarks

7 reported for Codestral-22B · 15 for Qwen2.5-Coder 32B Instruct

2 shared

Codestral-22B outperforms in 0 benchmarks, while Qwen2.5-Coder 32B Instruct is better at 2 benchmarks (HumanEval, MBPP).

Qwen2.5-Coder 32B Instruct significantly outperforms across most benchmarks.

Fri Aug 28 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

9.8B diff

Qwen2.5-Coder 32B Instruct has 9.8B more parameters than Codestral-22B, making it 44.1% larger.

Mistral AI
Codestral-22B
22.2Bparameters
Alibaba Cloud / Qwen Team
Qwen2.5-Coder 32B Instruct
32.0Bparameters
22.2B
Codestral-22B
32.0B
Qwen2.5-Coder 32B Instruct

Context Window

Maximum input and output token capacity

Only Qwen2.5-Coder 32B Instruct specifies input context (128,000 tokens). Only Qwen2.5-Coder 32B Instruct specifies output context (128,000 tokens).

Mistral AI
Codestral-22B
Input- tokens
Output- tokens
Alibaba Cloud / Qwen Team
Qwen2.5-Coder 32B Instruct
Input128,000 tokens
Output128,000 tokens
Fri Aug 28 2026 • llm-stats.com

License

Usage and distribution terms

Codestral-22B is licensed under MNPL-0.1, while Qwen2.5-Coder 32B Instruct uses Apache 2.0.

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

Codestral-22B

MNPL-0.1

Open weights

Qwen2.5-Coder 32B Instruct

Apache 2.0

Open weights

Release Timeline

When each model was launched

Codestral-22B was released on 2024-05-29, while Qwen2.5-Coder 32B Instruct was released on 2024-09-19.

Qwen2.5-Coder 32B Instruct is 4 months newer than Codestral-22B.

Codestral-22B

May 29, 2024

2.2 years ago

Qwen2.5-Coder 32B Instruct

Sep 19, 2024

1.9 years ago

3mo 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

Judge for yourself.

Run your own prompts against Codestral-22B and Qwen2.5-Coder 32B Instruct side-by-side, then vote on the output you prefer.

Codestral-22B
✓ Preferred
Qwen2.5-Coder 32B Instruct
Open in Playground

FAQ

Common questions about Codestral-22B vs Qwen2.5-Coder 32B Instruct.

Which is better, Codestral-22B or Qwen2.5-Coder 32B Instruct?

Codestral-22B and Qwen2.5-Coder 32B Instruct are closely matched on the LLM Stats Score at 0.5 and 2.7. Codestral-22B is made by Mistral AI and Qwen2.5-Coder 32B Instruct is made by Alibaba Cloud / Qwen Team. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Codestral-22B compare to Qwen2.5-Coder 32B Instruct in benchmarks?

Codestral-22B scores HumanEvalFIM-Average: 91.6%, HumanEval: 81.1%, MBPP: 78.2%, Spider: 63.5%, HumanEval-Average: 61.5%. Qwen2.5-Coder 32B Instruct scores HumanEval: 92.7%, GSM8k: 91.1%, MBPP: 90.2%, HellaSwag: 83.0%, Winogrande: 80.8%.

What are the context window sizes for Codestral-22B and Qwen2.5-Coder 32B Instruct?

Codestral-22B supports an unknown number of tokens and Qwen2.5-Coder 32B Instruct supports 128K 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 Qwen2.5-Coder 32B Instruct?

Key differences include LLM Stats Score (0.5 vs 2.7), licensing (MNPL-0.1 vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes Codestral-22B and Qwen2.5-Coder 32B Instruct?

Codestral-22B is developed by Mistral AI and Qwen2.5-Coder 32B Instruct is developed by Alibaba Cloud / Qwen Team.