Mistral Small 4 API Pricing: Europe's Best Budget Model at $0.15/M

Mistral Small 4 offers capable general-purpose AI at $0.15/M input, $0.60/M output — the cheapest general-purpose model from a European provider with GDPR-friendly infrastructure.

Updated Aug 10, 2026 · 94 models tracked across 11 providers

TL;DR

🇪🇺 European AI: Mistral is headquartered in Paris with European API infrastructure. For companies that need GDPR-compliant AI processing within EU jurisdiction, Mistral Small 4 is the cheapest general-purpose option from a European provider.

Mistral Small 4 Pricing Breakdown

At $0.15 per million input tokens and $0.60 per million output tokens, Mistral Small 4 sits in the budget tier — more capable than ultra-budget models (Qwen 3.7 Flash, GPT-5 nano) but significantly cheaper than mid-tier options (GPT-5.4, Claude Sonnet 5). Here's how costs scale:

Monthly Volume Input Cost Output Cost Total (50/50 split)
1M tokens$0.15$0.60$0.38
10M tokens$1.50$6.00$3.75
100M tokens$15.00$60.00$37.50
1B tokens$150.00$600.00$375.00

At 100M tokens/month, Mistral Small 4 costs $37.50 — about half the cost of Mistral Large 3 ($100) and a fraction of Claude Sonnet 5 ($600) or GPT-5.4 ($87.50).

Mistral Small 4 vs Other Budget Models

Model Input $/M Output $/M Context Provider Region
Qwen 3.7 Flash $0.03 $0.13 1M Alibaba China
GPT-5 nano $0.05 $0.40 128K OpenAI US
Gemini 2.5 Flash-Lite $0.10 $0.40 1M Google US
Ministral 3 3B $0.10 $0.10 128K Mistral EU
Mistral Small 4 $0.15 $0.60 128K Mistral EU
DeepSeek V4 Flash $0.14 $0.28 1M DeepSeek China
Devstral Small 2 $0.10 $0.30 128K Mistral EU
Mistral Large 3 $0.50 $1.50 262K Mistral EU

Mistral Small 4 is slightly more expensive than DeepSeek V4 Flash ($0.14/$0.28) and Ministral 3 3B ($0.10/$0.10), but it's a larger, more capable model. The key differentiator is European hosting — if you need EU data sovereignty, Mistral Small 4 is the cheapest general-purpose option.

Mistral Model Family: Which One?

Model Input $/M Output $/M Context Best For
Ministral 3 3B $0.10 $0.10 128K Ultra-budget, edge, output-heavy
Devstral Small 2 $0.10 $0.30 128K Budget coding tasks
Mistral Small 4 $0.15 $0.60 128K General-purpose budget
Magistral Small $0.50 $1.50 256K Reasoning tasks
Codestral $0.30 $0.90 256K Code generation specialist
Mistral Medium 3.5 $1.50 $7.50 128K Complex reasoning
Mistral Large 3 $0.50 $1.50 262K Best Mistral quality

Decision guide: Need the absolute cheapest? Ministral 3 3B. Need coding? Devstral Small 2 or Codestral. Need general-purpose quality on a budget? Mistral Small 4. Need the best Mistral quality? Mistral Large 3.

Best Use Cases for Mistral Small 4

🇪🇺 EU-Compliant Chatbots

Customer support and internal tools that need European data residency. Mistral's Paris-based infrastructure keeps data in the EU.

💻 Code Assistance

Code completion, review, and explanation at budget pricing. Good for developer tools that need quality without premium costs.

📄 Document Processing

Summarization, extraction, and analysis of business documents. 128K context handles most document lengths.

🤖 AI Agent Workflows

Multi-step agent chains where each step needs decent quality but cost control is important. Good price-to-capability ratio.

🎓 Education & Research

Research tools and educational platforms that need reliable language understanding at affordable per-token costs.

🔄 High-Volume Classification

Content moderation, intent detection, and sentiment analysis at scale. More capable than ultra-budget models for nuanced tasks.

Real-World Cost: 50K Support Conversations/Month

Consider a customer support chatbot handling 50,000 conversations per month. Each conversation averages 2,000 input tokens (conversation history + knowledge base context) and 500 output tokens (response).

Model Input Cost Output Cost Total/Month
Qwen 3.7 Flash $3.00 $3.25 $6.25
GPT-5 nano $5.00 $10.00 $15.00
DeepSeek V4 Flash $14.00 $7.00 $21.00
Mistral Small 4 $15.00 $15.00 $30.00
Mistral Large 3 $50.00 $37.50 $87.50
Claude Sonnet 5 $200.00 $250.00 $450.00

At $30/month for 50K conversations, Mistral Small 4 is 60% cheaper than Mistral Large 3 and 93% cheaper than Claude Sonnet 5 — while offering solid general-purpose quality for support use cases.

Compare Mistral Small 4 Pricing

See how Mistral Small 4 stacks up against 94 models across 11 providers.

Open Calculator Compare Models

Frequently Asked Questions

How much does Mistral Small 4 cost?
Mistral Small 4 costs $0.15 per million input tokens and $0.60 per million output tokens. The output-to-input ratio is 4:1 — typical for budget models. At 100M tokens/month with a 50/50 split, that's $37.50/month.
What is the context window of Mistral Small 4?
Mistral Small 4 has a 128K token context window. This is standard for budget models — sufficient for most chat, coding assistant, and document analysis tasks. For 1M+ context, consider Gemini 2.5 Flash-Lite ($0.10/$0.40) or GPT-5.6 Luna ($0.20/$1.20).
Is Mistral Small 4 cheaper than GPT-5 nano?
No. GPT-5 nano ($0.05/$0.40) is cheaper on both input (3x) and output (1.5x). However, Mistral Small 4 is a larger, more capable model — it handles complex reasoning and multi-step tasks better than GPT-5 nano. If you need quality on a budget, Mistral Small 4 is the better choice. If you need the absolute cheapest per-token cost, GPT-5 nano or Qwen 3.7 Flash ($0.03/$0.13) win.
Is Mistral Small 4 GDPR compliant?
Yes. Mistral is a French company headquartered in Paris, and their API infrastructure runs on European data centers. This makes Mistral Small 4 a strong choice for EU-based companies that need GDPR-compliant AI processing without data leaving European jurisdiction.
What is Mistral Small 4 good for?
Mistral Small 4 excels at general-purpose tasks: chatbots, content generation, code assistance, document summarization, and data extraction. It's a step up in capability from ultra-budget models (Qwen 3.7 Flash, GPT-5 nano) while remaining significantly cheaper than mid-tier models (GPT-5.4, Claude Sonnet 5).
How does Mistral Small 4 compare to Mistral Large 3?
Mistral Large 3 ($0.50/$1.50) costs 3.3x more on input and 2.5x more on output than Mistral Small 4 ($0.15/$0.60). Large 3 has a bigger context window (262K vs 128K) and stronger reasoning. Choose Small 4 for budget-sensitive workloads; choose Large 3 when you need Mistral's best quality.

Related Pages