FlagshipMistral

Mistral Medium 3.1

Mistral Medium 3.1 is Mistral's flagship multimodal model supporting text and image inputs with a 131K token context window.

Context 131K
Tier Flagship
Modalities text, image
Input from
$0.400 / 1M tokens
across 1 provider

API Pricing

ProviderInput / 1MOutput / 1MSpeedTTFTUpdated
$0.400$2.0098.0 t/s453ms4/14/2026

Prices updated daily. Last check: 4/14/2026

Model Details

General

Creator
Mistral
Family
Mistral
Tier
Flagship
Context Window
131K
Modalities
Text, Image

Capabilities

Tool Calling
No
Open Source
No

Strengths & Limitations

  • Multimodal support for both text and image inputs
  • Large 131,072 token context window for processing extensive documents
  • Fast generation speed at 94.53 tokens per second
  • Flagship-tier model representing Mistral's most capable offering
  • Relatively quick time to first token at 443 milliseconds
  • Strong performance for complex reasoning tasks
  • Suitable for sophisticated multimodal analysis workflows
  • No tool calling or function calling support
  • Proprietary model with no access to weights
  • Limited to text and image modalities (no audio or video)
  • Fewer API integrations compared to some competing flagship models
  • Higher resource requirements due to flagship positioning

Key Features

131,072 token context window
Text input processing
Image input support
Multimodal reasoning capabilities
Streaming response generation
Batch processing support
94.53 tokens per second generation speed
443ms time to first token

About Mistral Medium 3.1

Mistral Medium 3.1 is the flagship model from Mistral, representing the company's most capable offering in their model lineup. As a proprietary model, it sits at the top of Mistral's tier structure and competes with other flagship models in the market. The model features a 131,072 token context window and supports both text and image inputs, making it a multimodal solution for complex tasks. Performance benchmarks show it generates approximately 94.53 tokens per second with a time to first token of 443 milliseconds. However, the model does not include tool calling capabilities, which distinguishes it from some competing flagship models. Mistral Medium 3.1 is designed for applications requiring sophisticated reasoning across text and visual content. Organizations typically deploy it for complex analysis tasks that benefit from its multimodal capabilities and substantial context window, though the absence of tool calling may limit its use in agentic workflows compared to other flagship alternatives.

Common Use Cases

Mistral Medium 3.1 is well-suited for complex multimodal analysis tasks that require processing both text and visual content within a single context. Its large context window makes it effective for analyzing lengthy documents with embedded images, conducting detailed visual content analysis, and handling sophisticated reasoning tasks across multiple modalities. The model works well for research applications, content analysis, document processing with visual elements, and scenarios where high-quality multimodal understanding is needed. However, organizations requiring tool calling capabilities for agentic workflows may need to consider alternative flagship models or supplement with additional tooling.

Frequently Asked Questions

How much does Mistral Medium 3.1 cost per million tokens?

Mistral Medium 3.1 pricing varies by provider and pricing type (standard vs batch). Check the pricing table above for current rates across all providers.

What is Mistral Medium 3.1 best used for?

Mistral Medium 3.1 excels at complex multimodal tasks requiring analysis of both text and images. Its 131K token context window makes it particularly effective for processing lengthy documents with visual elements, conducting sophisticated reasoning across modalities, and handling research or analysis workflows that benefit from multimodal understanding.

Does Mistral Medium 3.1 support tool calling or function calling?

No, Mistral Medium 3.1 does not include tool calling or function calling capabilities. This limits its use in agentic workflows compared to some other flagship models that offer these features. Users requiring tool integration would need to implement external orchestration or consider alternative models.