Mistral Small 4
MistralBalanced
ThinkingTool UseVisionStructured Output
About this model
Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system. It combines strong reasoning from Magistral, multimodal understanding from Pixtral, and agentic coding capabilities from Devstral, enabling one model to handle complex analysis, software development, and visual tasks within the same workflow.
Performance Tier
Balanced
Mistral Small 4 is a balanced model from Mistral : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plans| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.150 |
| Output (completion) | $0.600 |
| Cache read | $0.015 |
Capabilities
Context Length262K
Max Output Tokens—
TokenizerMistral
Inputtext, image
Outputtext
Release DateMarch 16, 2026
Benchmarks
General Intelligence
MMLU
Not reported
GPQA Diamond
71.2%
Mathematics
MATH-500
Not reported
Programming
HumanEval
Not reported
Recommended Use Cases
General ChatCodingTranslationAnalysisCustomer Support
Strengths
- First Mistral model to unify reasoning (Magistral), vision (Pixtral), and coding (Devstral)
- Extremely cost-efficient — only 6B active parameters from 119B MoE (128 experts)
- 256K context window with configurable reasoning effort (none/low/medium/high)
- Open-weight Apache 2.0 license, EU-based — ideal for GDPR compliance
Limitations
- Very recent release — community benchmarks and evaluations still pending
- Outperformed by Mistral Large on complex reasoning tasks
Resources
This model may use your data for training
