Cracked AI Engineering

Mistral Small 3.2 24B Instruct (2506) vs Qwen3 235B A22B Instruct 2507

Qwen3 235B A22B Instruct 2507 offers 260K tokens context vs 128K tokens, Mistral Small 3.2 24B Instruct (2506) is more cost-effective, Mistral Small 3.2 24B Instruct (2506) supports vision. Compare full specs, pricing, and choose the best model for your use case.

Quick Overview

Mistral Small 3.2 24B Instruct (2506)

Scaleway

128K tokens context • $0.15 / $0.35 per 1M tokens

View full specifications →

Qwen3 235B A22B Instruct 2507

Scaleway

260K tokens context • $0.75 / $2.25 per 1M tokens

View full specifications →

Detailed Comparison

Specification
Mistral Small 3.2 24B Instruct (2506)
Qwen3 235B A22B Instruct 2507
Provider
Scaleway
Scaleway
Context Window
128K tokens
260K tokens
Max Output Tokens
8K tokens
8K tokens
Input Pricing (per 1M tokens)
$0.15
$0.75
Output Pricing (per 1M tokens)
$0.35
$2.25
Release Date
Jun 2025
Jul 2025

Capabilities

Capability
Mistral Small 3.2 24B Instruct (2506)
Qwen3 235B A22B Instruct 2507
Text Generation
Vision
Function Calling
File Attachments

Which Model Should You Choose?

Choose Mistral Small 3.2 24B Instruct (2506) if:

  • • Cost efficiency is a priority

Choose Qwen3 235B A22B Instruct 2507 if:

  • • You need a larger context window