The AI arena is free today

Open Superagent

Codestral-22B vs Qwen3-Coder 480B A35B Instruct

Qwen3-Coder 480B A35B Instruct leads the LLM Stats Score 20.8 to -0.3.

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

Which is better?

Qwen3-Coder 480B A35B Instruct leads the overall LLM Stats Score 20.8 to -0.3, ranking #213 overall.

In the 1 individual benchmarks reported for both models, Codestral-22B wins 1; 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 value its reported benchmark strengths — it wins 1 of 1 exact shared results

Choose Qwen3-Coder 480B A35B Instruct

  • overall performance matters — it scores 20.8 and ranks #213 on LLM Stats
  • your work emphasizes reasoning — it leads those capability indexes
  • you want the most recent training data — it shipped Jan 2025

At a glance

The differences that matter most.

Codestral-22B
Mistral AI
Qwen3-Coder 480B A35B Instruct
Alibaba Cloud / Qwen Team
Core performance indexes
LLM Stats Score
-0.3
#350
20.8
#213
Reasoning
-0.4
#343
20.9
#205
Coding
2.4
#248
12.1
#170
Cost, coverage & limits
Benchmark wins
1 of 1
0 of 1
Input price
— / M
$0.30 / M
Output price
— / M
$1.00 / M
Context window
—
262,144

Individual benchmarks

7 reported for Codestral-22B · 10 for Qwen3-Coder 480B A35B Instruct

1 shared

Codestral-22B outperforms in 1 benchmarks (Spider), while Qwen3-Coder 480B A35B Instruct is better at 0 benchmarks.

Codestral-22B significantly outperforms across most benchmarks.

Tue Oct 20 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

457.8B diff

Qwen3-Coder 480B A35B Instruct has 457.8B more parameters than Codestral-22B, making it 2062.2% larger.

Mistral AI
Codestral-22B
22.2Bparameters
Alibaba Cloud / Qwen Team
Qwen3-Coder 480B A35B Instruct
480.0Bparameters
22.2B
Codestral-22B
480.0B
Qwen3-Coder 480B A35B Instruct

Context Window

Maximum input and output token capacity

Only Qwen3-Coder 480B A35B Instruct specifies input context (262,144 tokens). Only Qwen3-Coder 480B A35B Instruct specifies output context (262,144 tokens).

Mistral AI
Codestral-22B
Input- tokens
Output- tokens
Alibaba Cloud / Qwen Team
Qwen3-Coder 480B A35B Instruct
Input262,144 tokens
Output262,144 tokens
Tue Oct 20 2026 • llm-stats.com

License

Usage and distribution terms

Codestral-22B is licensed under MNPL-0.1, while Qwen3-Coder 480B A35B 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

Qwen3-Coder 480B A35B Instruct

Apache 2.0

Open weights

Release Timeline

When each model was launched

Codestral-22B was released on 2024-05-29, while Qwen3-Coder 480B A35B Instruct was released on 2025-01-31.

Qwen3-Coder 480B A35B Instruct is 8 months newer than Codestral-22B.

Codestral-22B

May 29, 2024

2.4 years ago

Qwen3-Coder 480B A35B Instruct

Jan 31, 2025

1.7 years ago

8mo 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 Qwen3-Coder 480B A35B Instruct side-by-side, then vote on the output you prefer.

Codestral-22B
✓ Preferred
Qwen3-Coder 480B A35B Instruct
Open in Playground

FAQ

Common questions about Codestral-22B vs Qwen3-Coder 480B A35B Instruct.

Which is better, Codestral-22B or Qwen3-Coder 480B A35B Instruct?

Qwen3-Coder 480B A35B Instruct leads the LLM Stats Score 20.8 to -0.3. Codestral-22B is made by Mistral AI and Qwen3-Coder 480B A35B 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 Qwen3-Coder 480B A35B Instruct in benchmarks?

Codestral-22B scores HumanEvalFIM-Average: 91.6%, HumanEval: 81.1%, MBPP: 78.2%, Spider: 63.5%, HumanEval-Average: 61.5%. Qwen3-Coder 480B A35B Instruct scores TAU-bench Retail: 77.5%, SWE-Bench Verified: 69.6%, BFCL-v3: 68.7%, Aider-Polyglot: 61.8%, TAU-bench Airline: 60.0%.

What are the context window sizes for Codestral-22B and Qwen3-Coder 480B A35B Instruct?

Codestral-22B supports an unknown number of tokens and Qwen3-Coder 480B A35B Instruct supports 262K 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 Qwen3-Coder 480B A35B Instruct?

Key differences include LLM Stats Score (-0.3 vs 20.8), licensing (MNPL-0.1 vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes Codestral-22B and Qwen3-Coder 480B A35B Instruct?

Codestral-22B is developed by Mistral AI and Qwen3-Coder 480B A35B Instruct is developed by Alibaba Cloud / Qwen Team.

Related comparisons