Mistral Large
Description
Mistral Large is Mistral AI's flagship offering. The model excels at reasoning, code generation, JSON handling, and chat applications. It is a proprietary model with support for dozens of languages including French, German, Spanish, Italian, Portuguese, Arabic, Hindi, Russian, Chinese, Japanese, and Korean, along with 80+ coding languages.
Technical Specifications
| Specification |
Value |
| Context Window |
128,000 tokens |
| Input Modalities |
Text |
| Output Modalities |
Text |
| Architecture |
Transformer (proprietary) |
Pricing
| Type |
Price |
| Input Tokens |
$2.00 per 1M tokens |
| Output Tokens |
$6.00 per 1M tokens |
Capabilities
| Capability |
Status |
| Tool/Function Calling |
Supported |
| Structured Outputs |
Supported |
| JSON Mode |
Supported |
| Long-Context Processing |
Supported (128K) |
| Reasoning Tasks |
Supported |
| Multilingual |
Supported (dozens of languages) |
| Code Generation |
Supported (80+ languages) |
| Vision |
Not Supported |
Supported Parameters
| Parameter |
Supported |
max_tokens |
Yes |
temperature |
Yes |
top_p |
Yes |
stop |
Yes (stop sequences) |
frequency_penalty |
Yes |
presence_penalty |
Yes |
seed |
Yes |
response_format |
Yes |
structured_outputs |
Yes |
tools |
Yes |
tool_choice |
Yes |
API Usage Example
curl https://api.langmart.ai/v1/chat/completions \
-H "Authorization: Bearer $LANGMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "mistralai/mistral-large",
"messages": [
{
"role": "user",
"content": "Explain the concept of neural networks."
}
],
"temperature": 0.7,
"max_tokens": 1024
}'
Mistral AI Family
mistralai/mistral-small - Smaller, faster model for simpler tasks
mistralai/mistral-medium - Mid-tier model (deprecated)
mistralai/mistral-tiny - Lightweight model for basic tasks
mistralai/mixtral-8x7b - Mixture of Experts model
mistralai/mixtral-8x22b - Larger Mixture of Experts model
mistralai/codestral - Specialized for code generation
Comparable Models (Other Providers)
anthropic/claude-3-opus - Anthropic's flagship model
openai/gpt-4-turbo - OpenAI's advanced model
google/gemini-pro-1.5 - Google's flagship model
Model Identification
| Field |
Value |
| Model Name |
Mistral Large |
| Model ID |
mistralai/mistral-large |
| Version |
mistral-large-2407 |
| Author/Provider |
Mistral AI |
| Created |
February 26, 2024 |
| Headquarters |
France |
Language Support
Natural Languages
- French
- German
- Spanish
- Italian
- Portuguese
- Arabic
- Hindi
- Russian
- Chinese
- Japanese
- Korean
- English
- And more...
Programming Languages
Supports 80+ coding languages for code generation and understanding.
Mistral AI
| Field |
Value |
| Base URL |
https://api.mistral.ai/v1 |
| Status Page |
https://status.mistral.ai/ |
| Data Policy |
No training use; retains prompts for 30 days |
| Headquarters |
France |
Key Use Cases
- Reasoning & Analysis - Complex problem-solving and logical reasoning
- Code Generation - Writing and understanding code in 80+ languages
- JSON Handling - Structured data processing and generation
- Chat Applications - Conversational AI and assistants
- Long Document Processing - Handling documents up to 128K tokens
- Multilingual Tasks - Translation and cross-lingual applications
Notes
- Mistral Large represents Mistral AI's most capable model
- The 128K context window enables processing of long documents
- Tool calling support makes it suitable for agent-based applications
- Structured output support enables reliable JSON generation
- Data retention policy: prompts retained for 30 days (no training use)
Source: LangMart API - https://langmart.ai/model-docs
Last Updated: December 2024