Mistral Small is a lightweight model designed for cost-effective use in tasks like translation and summarization. Available under the Mistral Research License, this model offers customers a cost-efficient, fast, and reliable solution for tasks like translation, summarization, and sentiment analysis, without the need for full-scale general-purpose models.
With 22 billion parameters, Mistral Small v24.09 serves as a convenient middle ground between Mistral NeMo 12B and Mistral Large 2, providing a cost-effective option that can be deployed across various platforms and environments. The new version also demonstrates notable improvements in human alignment, reasoning, and code generation compared to the previous model.
Key Features:
- Cost-efficient: Strikes a balance between Mistral NeMo 12B and Mistral Large 2, catering to a variety of use cases.
- Versatile: Excels in tasks such as translation, summarization, and sentiment analysis.
- Flexible deployment: Can be implemented across diverse platforms and environments.
- Performance upgrade: Features notable improvements compared to the previous Mistral Small v24.02 model.
- Balanced solution: Provides a fast and dependable alternative without the need for full-scale general-purpose models.
- 128k sequence length
Run mistral-small on Ollama
Run mistral-small on Ollama:
ollama run mistral-small