NostraEmpire/mirror-mistral-7b-instruct-v0.3

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7BQuant:FP8Context Size:4kTool Calling:SupportedPublished:Aug 31, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

NostraEmpire/mirror-mistral-7b-instruct-v0.3 is an instruction-tuned large language model based on Mistral AI's 7 billion parameter Mistral-7B-v0.3 architecture. This model features an extended vocabulary, a v3 tokenizer, and notably supports function calling, enhancing its utility for structured interactions. It is designed for instruct-following tasks and can be integrated with both mistral-inference and Hugging Face transformers for various applications.

Loading preview...

Model Overview

NostraEmpire/mirror-mistral-7b-instruct-v0.3 is an instruction-tuned variant of Mistral AI's Mistral-7B-v0.3 large language model. This 7 billion parameter model builds upon its predecessor with several key enhancements, making it more versatile for interactive applications.

Key Capabilities

  • Extended Vocabulary: Features an expanded vocabulary to 32768 tokens, allowing for broader linguistic coverage.
  • V3 Tokenizer Support: Utilizes an updated v3 tokenizer for improved tokenization efficiency and quality.
  • Function Calling: A significant differentiator, this model supports function calling, enabling it to interact with external tools and APIs. This capability is demonstrated with examples for fetching weather information.
  • Instruct Following: Optimized for understanding and executing instructions, making it suitable for chat and command-based applications.

Good For

  • Interactive Chatbots: Its instruct-following and function calling capabilities make it well-suited for developing conversational agents that can perform actions.
  • Tool-Augmented LLM Applications: Developers can leverage its function calling feature to integrate the model with custom tools and services.
  • General Instruction-Based Tasks: Effective for a wide range of tasks requiring the model to follow specific directions or answer questions based on provided context.

Limitations

The model is a demonstration of fine-tuning capabilities and currently lacks built-in moderation mechanisms. Users should implement their own guardrails for deployments requiring moderated outputs.

Finetunes
1 models